diff options
26 files changed, 2600 insertions, 2374 deletions
diff --git a/doc/context/scripts/mkiv/mtx-bibtex.html b/doc/context/scripts/mkiv/mtx-bibtex.html index ba1591b4b..61a679716 100644 --- a/doc/context/scripts/mkiv/mtx-bibtex.html +++ b/doc/context/scripts/mkiv/mtx-bibtex.html @@ -41,12 +41,14 @@ <tr><th/><td/><td/></tr> <tr><th>--toxml</th><td></td><td>convert bibtex database(s) to xml</td></tr> <tr><th>--tolua</th><td></td><td>convert bibtex database(s) to lua</td></tr> + <tr><th>--search</th><td></td><td>seatch bibtex database(s)</td></tr> </table> <br/> <h1>Example</h1> <tt>mtxrun --script bibtex --tolua bibl-001.bib</tt> <br/><tt>mtxrun --script bibtex --tolua --simple bibl-001.bib</tt> <br/><tt>mtxrun --script bibtex --toxml bibl-001.bib bibl-002.bib bibl-003.bib biblio.xml</tt> +<br/><tt>mtxrun --script bibtex --search --list --pattern=match(author:foo) bar.bib</tt> <br/><br/> </div> </div> </body> diff --git a/doc/context/scripts/mkiv/mtx-bibtex.man b/doc/context/scripts/mkiv/mtx-bibtex.man index cedf41b8b..ef44fc87d 100644 --- a/doc/context/scripts/mkiv/mtx-bibtex.man +++ b/doc/context/scripts/mkiv/mtx-bibtex.man @@ -16,6 +16,9 @@ convert bibtex database(s) to xml .TP .B --tolua convert bibtex database(s) to lua +.TP +.B --search +seatch bibtex database(s) .SH AUTHOR More information about ConTeXt and the tools that come with it can be found at: diff --git a/doc/context/scripts/mkiv/mtx-bibtex.xml b/doc/context/scripts/mkiv/mtx-bibtex.xml index b33e1809c..5c28b946a 100644 --- a/doc/context/scripts/mkiv/mtx-bibtex.xml +++ b/doc/context/scripts/mkiv/mtx-bibtex.xml @@ -10,6 +10,7 @@ <subcategory> <flag name="toxml"><short>convert bibtex database(s) to xml</short></flag> <flag name="tolua"><short>convert bibtex database(s) to lua</short></flag> + <flag name="search"><short>seatch bibtex database(s)</short></flag> </subcategory> </category> </flags> @@ -20,6 +21,7 @@ <example><command>mtxrun --script bibtex --tolua bibl-001.bib</command></example> <example><command>mtxrun --script bibtex --tolua --simple bibl-001.bib</command></example> <example><command>mtxrun --script bibtex --toxml bibl-001.bib bibl-002.bib bibl-003.bib biblio.xml</command></example> + <example><command>mtxrun --script bibtex --search --list --pattern=match(author:foo) bar.bib</command></example> </subcategory> </category> </examples> diff --git a/scripts/context/lua/mtx-bibtex.lua b/scripts/context/lua/mtx-bibtex.lua index c81fd596f..92036e3a5 100644 --- a/scripts/context/lua/mtx-bibtex.lua +++ b/scripts/context/lua/mtx-bibtex.lua @@ -19,6 +19,7 @@ local helpinfo = [[ <subcategory> <flag name="toxml"><short>convert bibtex database(s) to xml</short></flag> <flag name="tolua"><short>convert bibtex database(s) to lua</short></flag> + <flag name="search"><short>seatch bibtex database(s)</short></flag> </subcategory> </category> </flags> @@ -29,6 +30,7 @@ local helpinfo = [[ <example><command>mtxrun --script bibtex --tolua bibl-001.bib</command></example> <example><command>mtxrun --script bibtex --tolua --simple bibl-001.bib</command></example> <example><command>mtxrun --script bibtex --toxml bibl-001.bib bibl-002.bib bibl-003.bib biblio.xml</command></example> + <example><command>mtxrun --script bibtex --search --list --pattern=match(author:foo) bar.bib</command></example> </subcategory> </category> </examples> @@ -43,7 +45,9 @@ local application = logs.application { local report = application.report +require("util-seq") require("publ-dat") +require("publ-fnd") scripts = scripts or { } scripts.bibtex = scripts.bibtex or { } @@ -57,7 +61,7 @@ function scripts.bibtex.toxml(files) if filetype == "xml" then target = filename elseif filetype == "bib" then - bibtex.load(instance,filename) + bibtex.load { dataset = instance, filename = filename } else -- not supported end @@ -77,7 +81,8 @@ function scripts.bibtex.tolua(files) if filetype == "lua" then target = filename elseif filetype == "bib" then - bibtex.load(instance,filename) + bibtex.load { dataset = instance, filename = filename } + else -- not supported end @@ -92,7 +97,48 @@ function scripts.bibtex.tolua(files) end end -if environment.arguments.toxml then +function scripts.bibtex.search(files,pattern,list) + if pattern then + local dataset = publications.datasets["whatever"] + for i=1,#files do + local filename = resolvers.findfile(files[i]) + if filename and filename ~= "" then + publications.load { dataset = "whatever", filename = filename } + end + end + local found = publications.search(dataset,pattern) + local tags = table.sortedkeys(found) + if #tags == 0 then + report("no match") + elseif list then + report("%s matches:",#tags) + local result = { } + local luadata = dataset.luadata + for i=1,#tags do + local tag = tags[i] + local entry = luadata[tag] + result[i] = { + tag, + entry.year, + entry.author, + entry.title, + } + end + utilities.formatters.formatcolumns(result) + logs.newline() + for i=1,#result do + texio.write_nl(result[i]) + end + logs.newline() + else + report("%s matches: % t",#tags,tags) + end + end +end + +if environment.arguments.search then + scripts.bibtex.search(environment.files,environment.arguments.pattern,environment.arguments.list) +elseif environment.arguments.toxml then scripts.bibtex.toxml(environment.files) elseif environment.arguments.tolua then scripts.bibtex.tolua(environment.files) diff --git a/tex/context/base/cont-new.mkiv b/tex/context/base/cont-new.mkiv index e3ce4e077..dcabceb93 100644 --- a/tex/context/base/cont-new.mkiv +++ b/tex/context/base/cont-new.mkiv @@ -11,7 +11,7 @@ %C therefore copyrighted by \PRAGMA. See mreadme.pdf for %C details. -\newcontextversion{2014.11.11 12:12} +\newcontextversion{2014.11.12 11:56} %D This file is loaded at runtime, thereby providing an excellent place for %D hacks, patches, extensions and new features. diff --git a/tex/context/base/context-version.pdf b/tex/context/base/context-version.pdf Binary files differindex 2391c5443..5c4a0262d 100644 --- a/tex/context/base/context-version.pdf +++ b/tex/context/base/context-version.pdf diff --git a/tex/context/base/context.mkiv b/tex/context/base/context.mkiv index 179709972..d625767ee 100644 --- a/tex/context/base/context.mkiv +++ b/tex/context/base/context.mkiv @@ -28,7 +28,7 @@ %D up and the dependencies are more consistent. \edef\contextformat {\jobname} -\edef\contextversion{2014.11.11 12:12} +\edef\contextversion{2014.11.12 11:56} \edef\contextkind {beta} %D For those who want to use this: diff --git a/tex/context/base/core-con.lua b/tex/context/base/core-con.lua index dd9f50dc8..e805b23e9 100644 --- a/tex/context/base/core-con.lua +++ b/tex/context/base/core-con.lua @@ -60,6 +60,14 @@ local counters = allocate { 0x006F, 0x0070, 0x0072, 0x0073, 0x0161, 0x0074, 0x0075, 0x0076, 0x007A, 0x017E }, + ['spanish'] = { + 0x0061, 0x0062, 0x0063, 0x0064, 0x0065, + 0x0066, 0x0067, 0x0068, 0x0069, 0x006A, + 0x006B, 0x006C, 0x006D, 0x006E, 0x00F1, + 0x006F, 0x0070, 0x0071, 0x0072, 0x0073, + 0x0074, 0x0075, 0x0076, 0x0077, 0x0078, + 0x0079, 0x007A + }, ['greek'] = { -- this should be the lowercase table -- 0x0391, 0x0392, 0x0393, 0x0394, 0x0395, -- 0x0396, 0x0397, 0x0398, 0x0399, 0x039A, @@ -131,6 +139,7 @@ counters['ar'] = counters['arabic'] counters['gr'] = counters['greek'] counters['g'] = counters['greek'] counters['sl'] = counters['slovenian'] +counters['es'] = counters['spanish'] counters['kr'] = counters['korean'] counters['kr-p'] = counters['korean-parent'] counters['kr-c'] = counters['korean-circle'] @@ -191,6 +200,8 @@ converters.maxchrs = maxchrs local lowercharacter = characters.lcchars local uppercharacter = characters.ucchars +local defaultcounter = counters.default + local function do_alphabetic(n,mapping,mapper,t) -- todo: make zero based variant (initial n + 1) if not t then t = { } @@ -208,11 +219,11 @@ local function do_alphabetic(n,mapping,mapper,t) -- todo: make zero based varian end function converters.alphabetic(n,code) - return do_alphabetic(n,counters[code] or counters.default,lowercharacter) + return do_alphabetic(n,code and counters[code] or defaultcounter,lowercharacter) end function converters.Alphabetic(n,code) - return do_alphabetic(n,counters[code] or counters.default,uppercharacter) + return do_alphabetic(n,code and counters[code] or defaultcounter,uppercharacter) end local lower_offset = 96 @@ -228,8 +239,8 @@ converters['A'] = converters.Characters converters['AK'] = converters.Characters converters['KA'] = converters.Characters -function commands.alphabetic(n,c) context(do_alphabetic(n,counters[c],lowercharacter)) end -function commands.Alphabetic(n,c) context(do_alphabetic(n,counters[c],uppercharacter)) end +function commands.alphabetic(n,c) context(do_alphabetic(n,c and counters[c] or defaultcounter,lowercharacter)) end +function commands.Alphabetic(n,c) context(do_alphabetic(n,c and counters[c] or defaultcounter,uppercharacter)) end function commands.character (n) context(chr (n,lower_offset)) end function commands.Character (n) context(chr (n,upper_offset)) end function commands.characters(n) context(chrs(n,lower_offset)) end @@ -893,7 +904,7 @@ local words = { [900] = "novecientos", [1000] = "mil", [1000^2] = "millón", - [1000^3] = "mil millónes", + [1000^3] = "mil millones", [1000^4] = "billón", } diff --git a/tex/context/base/core-con.mkiv b/tex/context/base/core-con.mkiv index 47c7c88d5..34ab90839 100644 --- a/tex/context/base/core-con.mkiv +++ b/tex/context/base/core-con.mkiv @@ -79,6 +79,9 @@ \def\languagecharacters#1{\ctxcommand{alphabetic(\number#1,"\currentlanguage")}} % new \def\languageCharacters#1{\ctxcommand{Alphabetic(\number#1,"\currentlanguage")}} % new +\def\alphabeticnumerals#1{\ctxcommand{alphabetic(\number#1)}} +\def\Alphabeticnumerals#1{\ctxcommand{Alphabetic(\number#1)}} + % we could use an auxiliary macro to save some bytes in the format % % \def\dolanguagecharacters#1#2{\ctxcommand{alphabetic(\number#2,"#1")}} @@ -627,6 +630,9 @@ \defineconversion [AK] [\smallcappedcharacters] \defineconversion [KA] [\smallcappedcharacters] +\defineconversion [\v!alphabetic] [\alphabeticnumerals] +\defineconversion [\v!Alphabetic] [\Alphabeticnumerals] + \defineconversion [\v!number] [\numbers] \defineconversion [\v!numbers] [\numbers] \defineconversion [\v!Numbers] [\Numbers] @@ -659,6 +665,9 @@ \defineconversion [\v!greek] [\greeknumerals] \defineconversion [\v!Greek] [\Greeknumerals] +\defineconversion [ñ] [\spanishnumerals] +\defineconversion [Ñ] [\Spanishnumerals] + \defineconversion [g] [\greeknumerals] \defineconversion [G] [\Greeknumerals] diff --git a/tex/context/base/lang-def.mkiv b/tex/context/base/lang-def.mkiv index 088f86eb8..2d9c88a18 100644 --- a/tex/context/base/lang-def.mkiv +++ b/tex/context/base/lang-def.mkiv @@ -271,11 +271,6 @@ \installlanguage [\s!slovenian] [\s!sl] \installlanguage [slovene] [\s!sl] % both possible (mojca: still needed?) -\def\doconvertsloveniancharacters{\dodoconvertcharacters{25}} - -\def\sloveniancharacters{\doconvertsloveniancharacters\sloveniancharacter} -\def\slovenianCharacters{\doconvertsloveniancharacters\slovenianCharacter} - %D Define general-purpose macros for Slovenian character enumerations: \defineconversion [sloveniancharacter] [\sloveniancharacter] @@ -284,13 +279,24 @@ \defineconversion [sloveniancharacters] [\sloveniancharacters] \defineconversion [slovenianCharacters] [\slovenianCharacters] +\defineconversion [spanishcharacter] [\spanishcharacter] +\defineconversion [spanishCharacter] [\spanishCharacter] + +\defineconversion [spanishcharacters] [\spanishcharacters] +\defineconversion [spanishCharacters] [\spanishCharacters] + %D Define these as the general character enumeration when %D language is Slovenian. If you feel uncomfortable with this, %D mail Mojca, since she promised to to take the heat. -\defineconversion [\s!sl] [character] [\sloveniancharacter] -\defineconversion [\s!sl] [Character] [\slovenianCharacter] +\def\sloveniancharacters#1{\ctxcommand{alphabetic(\number#1,"sl")}} +\def\slovenianCharacters#1{\ctxcommand{Alphabetic(\number#1,"sl")}} + +\def\spanishcharacters #1{\ctxcommand{alphabetic(\number#1,"es")}} +\def\spanishCharacters #1{\ctxcommand{Alphabetic(\number#1,"es")}} +\defineconversion [\s!sl] [character] [\sloveniancharacters] +\defineconversion [\s!sl] [Character] [\slovenianCharacters] \defineconversion [\s!sl] [characters] [\sloveniancharacters] \defineconversion [\s!sl] [Characters] [\slovenianCharacters] @@ -299,8 +305,15 @@ \defineconversion [\s!sl] [AK] [\smallcapped\sloveniancharacters] \defineconversion [\s!sl] [KA] [\smallcapped\sloveniancharacters] -\def\sloveniancharacters#1{\ctxcommand{alphabetic(\number#1,"sl")}} -\def\slovenianCharacters#1{\ctxcommand{Alphabetic(\number#1,"sl")}} +\defineconversion [\s!es] [character] [\spanishcharacters] +\defineconversion [\s!es] [Character] [\spanishCharacters] +\defineconversion [\s!es] [characters] [\spanishcharacters] +\defineconversion [\s!es] [Characters] [\spanishCharacters] + +\defineconversion [\s!es] [a] [\spanishcharacters] +\defineconversion [\s!es] [A] [\spanishCharacters] +\defineconversion [\s!es] [AK] [\smallcapped\spanishcharacters] +\defineconversion [\s!es] [KA] [\smallcapped\spanishcharacters] % Cyrillic Languages diff --git a/tex/context/base/mult-def.mkiv b/tex/context/base/mult-def.mkiv index 560faf6e0..7791200f9 100644 --- a/tex/context/base/mult-def.mkiv +++ b/tex/context/base/mult-def.mkiv @@ -34,6 +34,9 @@ % start todo in muld-def.lua: +\def\v!alphabetic {alphabetic} +\def\v!Alphabetic {Alphabetic} + \def\c!svgstyle {svgstyle} \def\c!nextleft {nextleft} diff --git a/tex/context/base/publ-aut.lua b/tex/context/base/publ-aut.lua index 7eef49f72..a449e25b4 100644 --- a/tex/context/base/publ-aut.lua +++ b/tex/context/base/publ-aut.lua @@ -27,6 +27,8 @@ local publications = publications local datasets = publications.datasets local writers = publications.writers local authors = publications.authors +local detailed = publications.detailed +local casters = publications.casters local chardata = characters.data @@ -210,6 +212,7 @@ local function splitauthorstring(str) end authors.splitstring = splitauthorstring +casters.author = splitauthorstring local function the_initials(initials,symbol,connector) if not symbol then @@ -307,18 +310,22 @@ function commands.btxauthorfield(i,field) end function commands.btxauthor(dataset,tag,field,settings) - local ds = datasets[dataset] - if not ds then + local current = datasets[dataset] + if not current then return f_invalid("dataset",dataset) end - local dt = ds.details[tag] - if not dt then - return f_invalid("details",tag) + local entry = current.luadata[tag] + if not entry then + return f_invalid("entry",tag) end - local split = dt[field] - if not split then + local value = entry[field] + if not value then return f_invalid("field",field) end + local split = detailed.author[value] + if type(split) ~= "table" then + return f_invalid("cast",value) + end local max = split and #split or 0 if max == 0 then return @@ -410,7 +417,6 @@ local function components(snippet,short) local initials = snippet.initials local firstnames = not short and snippet.firstnames local juniors = snippet.juniors --- inspect(initials) return vons and #vons > 0 and concat(vons, " ") or "", surnames and #surnames > 0 and concat(surnames, " ") or "", @@ -463,32 +469,26 @@ local default = { "author" } function authors.getauthor(dataset,tag,categories) local current = datasets[dataset] local luadata = current.luadata - if not luadata then - report("invalid dataset %a",dataset) - end - local entry = luadata[tag] + local entry = luadata and luadata[tag] if entry then local category = entry.category - local detail = current.details[tag] - if detail then - local list - if categories then - local c = categories[category] - if c then - local sets = c.sets - list = sets and sets.author and sets.authors or default - else - list = default - end + local list + if categories then + local c = categories[category] + if c then + local sets = c.sets + list = sets and sets.author and sets.authors or default else list = default end - for i=1,#list do - local l = list[i] - local d = detail[l] - if d then - return d, l - end + else + list = default + end + for i=1,#list do + local l = list[i] + local v = entry[l] + if v then + return detailed.author[v], l end end end @@ -510,23 +510,30 @@ end -- first : key author editor publisher title journal volume number pages -- second: year suffix title month day journal volume number +local function directget(dataset,entry,field) + local value = entry[field] + if value then + return detailed.author[value] + end +end + local function byauthor(dataset,list,method) - local luadata = datasets[dataset].luadata - local details = datasets[dataset].details + local current = datasets[dataset] + local luadata = current.luadata local result = { } local splitted = newsplitter(splitter) -- saves mem local snippets = { } -- saves mem + local get = publications.directget or directget + local field = "author" -- todo for i=1,#list do -- either { tag, tag, ... } or { { tag, index }, { tag, index } } local li = list[i] local tag = type(li) == "string" and li or li[1] - local entry = luadata[tag] - local detail = details[tag] local index = tostring(i) - if entry and detail then --- todo pluggable - local mainkey = writer(detail.author or detail.editor or entry.publisher or entry.title or "",snippets) - -- we could store the mainkey in details for tracing + local entry = luadata[tag] + if entry then + local value = get(current,entry,field) or "" + local mainkey = writer(value,snippets) result[i] = { index = i, split = { @@ -582,18 +589,3 @@ function authors.sorted(dataset,list,sorttype) -- experimental return valid end end - --- local dataset = publications.datasets.test --- --- local function add(str) --- dataset.details[str] = { author = publications.authors.splitstring(str) } --- end --- --- add("Hagen, Hans and Hoekwater, Taco Whoever T. Ex. and Henkel Hut, Hartmut Harald von der") --- add("Hans Hagen and Taco Whoever T. Ex. Hoekwater and Hartmut Harald von der Henkel Hut") --- add("de Gennes, P. and Gennes, P. de") --- add("van't Hoff, J. H. and {van't Hoff}, J. H.") --- --- local list = table.keys(dataset.details) --- local sort = publications.authors.sorted("test",list,"author") --- local test = { } for i=1,#sort do test[i] = dataset.details[list[sort[i]]] end diff --git a/tex/context/base/publ-dat.lua b/tex/context/base/publ-dat.lua index 0026c2c1f..b11ddf215 100644 --- a/tex/context/base/publ-dat.lua +++ b/tex/context/base/publ-dat.lua @@ -9,6 +9,7 @@ if not modules then modules = { } end modules ['publ-dat'] = { -- todo: strip the @ in the lpeg instead of on do_definition and do_shortcut -- todo: store bibroot and bibrootdt -- todo: dataset = datasets[dataset] => current = datasets[dataset] +-- todo: maybe split this file --[[ldx-- <p>This is a prelude to integrated bibliography support. This file just loads @@ -68,15 +69,31 @@ publications.tables = tables publications.statistics = publications.statistics or { } local publicationsstats = publications.statistics +local loaders = publications.loaders or { } +publications.loaders = loaders + +local casters = { } +publications.casters = casters + +local enhancers = publications.enhancers or { } +publications.enhancers = enhancers + +local enhancer = publications.enhancer or utilities.sequencers.new { arguments = "dataset" } +publications.enhancer = enhancer + +utilities.sequencers.appendgroup(enhancer,"system") -- private + publicationsstats.nofbytes = 0 publicationsstats.nofdefinitions = 0 publicationsstats.nofshortcuts = 0 publicationsstats.nofdatasets = 0 local privates = allocate { - category = true, - tag = true, - index = true, + category = true, + tag = true, + index = true, + suffix = true, + specification = true, } local specials = allocate { @@ -87,39 +104,6 @@ local specials = allocate { comment = true, } -tables.privates = privates -tables.specials = specials - -if not publications.usedentries then - function publications.usedentries() - return { } - end -end - -local v_all = interfaces and interfaces.variables.all or "all" - -local xmlplaceholder = "<?xml version='1.0' standalone='yes'?>\n<bibtex></bibtex>" - -local defaultshortcuts = allocate { - jan = "1", - feb = "2", - mar = "3", - apr = "4", - may = "5", - jun = "6", - jul = "7", - aug = "8", - sep = "9", - oct = "10", - nov = "11", - dec = "12", -} - -local space = p_whitespace^0 -local separator = space * "+" * space -local l_splitter = lpeg.tsplitat(separator) -local d_splitter = lpeg.splitat (separator) - local implicits = allocate { category = "implicit", tag = "implicit", @@ -148,12 +132,50 @@ local defaulttypes = allocate { author = "author", editor = "author", publisher = "author", + page = "pagenumber", + pages = "pagenumber", + keywords = "keyword", } tables.implicits = implicits tables.origins = origins tables.virtuals = virtuals tables.types = defaulttypes +tables.privates = privates +tables.specials = specials + +local variables = interfaces and interfaces.variables or setmetatableindex("self") + +local v_all = variables.all +local v_standard = variables.standard + +if not publications.usedentries then + function publications.usedentries() + return { } + end +end + +local xmlplaceholder = "<?xml version='1.0' standalone='yes'?>\n<bibtex></bibtex>" + +local defaultshortcuts = allocate { + jan = "1", + feb = "2", + mar = "3", + apr = "4", + may = "5", + jun = "6", + jul = "7", + aug = "8", + sep = "9", + oct = "10", + nov = "11", + dec = "12", +} + +local space = p_whitespace^0 +local separator = space * "+" * space +local l_splitter = lpeg.tsplitat(separator) +local d_splitter = lpeg.splitat (separator) local unknownfield = function(t,k) local v = "extra" @@ -280,7 +302,8 @@ function publications.parenttag(dataset,tag) elseif find(tag,"%+") then local tags = lpegmatch(l_splitter,tag) local parent = tags[1] - local luadata = datasets[dataset].luadata + local current = datasets[dataset] + local luadata = current.luadata local first = luadata[parent] if first then local combined = first.combined @@ -318,8 +341,8 @@ function publications.new(name) luadata = { }, suffixes = { }, xmldata = xmlconvert(xmlplaceholder), - -- details = { }, - -- ordered = { }, + details = { }, + ordered = { }, nofbytes = 0, entries = nil, -- empty == all sources = { }, @@ -332,39 +355,16 @@ function publications.new(name) resources = false, userdata = false, }, + specifications = { + -- used specifications + }, } -- we delay details till we need it (maybe we just delay the -- individual fields but that is tricky as there can be some -- depedencies) - setmetatableindex(dataset,function(t,k) - -- will become a plugin - if k == "details" then - if publications.enhance then - dataset.details = { } - publications.enhance(dataset.name) - return dataset.details - end - elseif k == "ordered" then - local luadata = dataset.luadata - local ordered = sortedkeys(luadata) - for i=1,#ordered do - ordered[i] = luadata[ordered[i]] - end - dataset.ordered = ordered - return ordered - end - end) return dataset end -function publications.markasupdated(name) - if type(name) == "string" then - rawset(datasets[name],"details",nil) - else - rawset(name,"details",nil) - end -end - setmetatableindex(datasets,function(t,k) if type(k) == "table" then return k -- so we can use this accessor as checker @@ -375,40 +375,6 @@ setmetatableindex(datasets,function(t,k) end end) --- we apply some normalization - -local space = S(" \t\n\r\f") -- / " " - ------ command = P("\\") * Cc("btxcmd{") * (R("az","AZ")^1) * Cc("}") ------ command = P("\\") * (Carg(1) * C(R("az","AZ")^1) / function(list,c) list[c] = (list[c] or 0) + 1 return "btxcmd{" .. c .. "}" end) -local command = P("\\") * (Carg(1) * C(R("az","AZ")^1) * space^0 / function(list,c) list[c] = (list[c] or 0) + 1 return "btxcmd{" .. c .. "}" end) -local somemath = P("$") * ((1-P("$"))^1) * P("$") -- let's not assume nested math -local any = P(1) -local done = P(-1) -local one_l = P("{") / "" -local one_r = P("}") / "" -local two_l = P("{{") / "" -local two_r = P("}}") / "" -local special = P("#") / "\\letterhash" - -local filter_0 = S('\\{}') -local filter_1 = (1-filter_0)^0 * filter_0 -local filter_2 = Cs( --- {{...}} ... {{...}} --- two_l * (command + special + any - two_r - done)^0 * two_r * done + --- one_l * (command + special + any - one_r - done)^0 * one_r * done + - (somemath + command + special + any )^0 -) - --- Currently we expand shortcuts and for large ones (like the acknowledgements --- in tugboat.bib) this is not that efficient. However, eventually strings get --- hashed again. - -local function do_shortcut(key,value,dataset) - publicationsstats.nofshortcuts = publicationsstats.nofshortcuts + 1 - dataset.shortcuts[key] = value -end - local function getindex(dataset,luadata,tag) local found = luadata[tag] if found then @@ -422,480 +388,517 @@ end publications.getindex = getindex --- todo: categories : metatable that lowers and also counts --- todo: fields : metatable that lowers - -local tags = table.setmetatableindex("table") - -local function do_definition(category,tag,tab,dataset) - publicationsstats.nofdefinitions = publicationsstats.nofdefinitions + 1 - local fields = dataset.fields - local luadata = dataset.luadata - if luadata[tag] then - local t = tags[tag] - local d = dataset.name - local n = (t[n] or 0) + 1 - t[d] = n - if trace_duplicates then - local p = { } - for k, v in sortedhash(t) do - p[#p+1] = formatters["%s:%s"](k,v) - end - report_duplicates("tag %a is present multiple times: % t",tag,p) - end - else - local found = luadata[tag] - local index = getindex(dataset,luadata,tag) - local entries = { - category = lower(category), - tag = tag, - index = index, - } - for i=1,#tab,2 do - local original = tab[i] - local normalized = fields[original] - if not normalized then - normalized = lower(original) -- we assume ascii fields - fields[original] = normalized - end - -- if entries[normalized] then - if rawget(entries,normalized) then - if trace_duplicates then - report_duplicates("redundant field %a is ignored for tag %a in dataset %a",normalized,tag,dataset.name) +do + + -- we apply some normalization + + local space = S(" \t\n\r\f") -- / " " + + ----- command = P("\\") * Cc("btxcmd{") * (R("az","AZ")^1) * Cc("}") + ----- command = P("\\") * (Carg(1) * C(R("az","AZ")^1) / function(list,c) list[c] = (list[c] or 0) + 1 return "btxcmd{" .. c .. "}" end) + local command = P("\\") * (Carg(1) * C(R("az","AZ")^1) * space^0 / function(list,c) list[c] = (list[c] or 0) + 1 return "btxcmd{" .. c .. "}" end) + local somemath = P("$") * ((1-P("$"))^1) * P("$") -- let's not assume nested math + local any = P(1) + local done = P(-1) + local one_l = P("{") / "" + local one_r = P("}") / "" + local two_l = P("{{") / "" + local two_r = P("}}") / "" + local special = P("#") / "\\letterhash" + + local filter_0 = S('\\{}') + local filter_1 = (1-filter_0)^0 * filter_0 + local filter_2 = Cs( + -- {{...}} ... {{...}} + -- two_l * (command + special + any - two_r - done)^0 * two_r * done + + -- one_l * (command + special + any - one_r - done)^0 * one_r * done + + (somemath + command + special + any )^0 + ) + + -- Currently we expand shortcuts and for large ones (like the acknowledgements + -- in tugboat.bib) this is not that efficient. However, eventually strings get + -- hashed again. + + local function do_shortcut(key,value,dataset) + publicationsstats.nofshortcuts = publicationsstats.nofshortcuts + 1 + dataset.shortcuts[key] = value + end + + -- todo: categories : metatable that lowers and also counts + -- todo: fields : metatable that lowers + + local tags = table.setmetatableindex("table") + + local function do_definition(category,tag,tab,dataset) + publicationsstats.nofdefinitions = publicationsstats.nofdefinitions + 1 + local fields = dataset.fields + local luadata = dataset.luadata + if luadata[tag] then + local t = tags[tag] + local d = dataset.name + local n = (t[n] or 0) + 1 + t[d] = n + if trace_duplicates then + local p = { } + for k, v in sortedhash(t) do + p[#p+1] = formatters["%s:%s"](k,v) end - else - local value = tab[i+1] - value = textoutf(value) - if lpegmatch(filter_1,value) then - value = lpegmatch(filter_2,value,1,dataset.commands) -- we need to start at 1 for { } + report_duplicates("tag %a is present multiple times: % t",tag,p) + end + else + local found = luadata[tag] + local index = getindex(dataset,luadata,tag) + local entries = { + category = lower(category), + tag = tag, + index = index, + } + for i=1,#tab,2 do + local original = tab[i] + local normalized = fields[original] + if not normalized then + normalized = lower(original) -- we assume ascii fields + fields[original] = normalized end - if normalized == "crossref" then - local parent = luadata[value] - if parent then - setmetatableindex(entries,parent) - else - -- warning + -- if entries[normalized] then + if rawget(entries,normalized) then + if trace_duplicates then + report_duplicates("redundant field %a is ignored for tag %a in dataset %a",normalized,tag,dataset.name) + end + else + local value = tab[i+1] + value = textoutf(value) + if lpegmatch(filter_1,value) then + value = lpegmatch(filter_2,value,1,dataset.commands) -- we need to start at 1 for { } + end + if normalized == "crossref" then + local parent = luadata[value] + if parent then + setmetatableindex(entries,parent) + else + -- warning + end end + entries[normalized] = value end - entries[normalized] = value end + luadata[tag] = entries end - luadata[tag] = entries end -end -local function resolve(s,dataset) - return dataset.shortcuts[s] or defaultshortcuts[s] or s -- can be number -end + local function resolve(s,dataset) + return dataset.shortcuts[s] or defaultshortcuts[s] or s -- can be number + end -local pattern = p_whitespace^0 - * C(P("message") + P("warning") + P("error") + P("comment")) * p_whitespace^0 * P(":") - * p_whitespace^0 - * C(P(1)^1) - -local function do_comment(s,dataset) - local how, what = lpegmatch(pattern,s) - if how and what then - local t = string.splitlines(utilities.strings.striplines(what)) - local b = file.basename(dataset.fullname or dataset.name or "unset") - for i=1,#t do - report("%s > %s : %s",b,how,t[i]) + local pattern = p_whitespace^0 + * C(P("message") + P("warning") + P("error") + P("comment")) * p_whitespace^0 * P(":") + * p_whitespace^0 + * C(P(1)^1) + + local function do_comment(s,dataset) + local how, what = lpegmatch(pattern,s) + if how and what then + local t = string.splitlines(utilities.strings.striplines(what)) + local b = file.basename(dataset.fullname or dataset.name or "unset") + for i=1,#t do + report("%s > %s : %s",b,how,t[i]) + end end end -end -local percent = P("%") -local start = P("@") -local comma = P(",") -local hash = P("#") -local escape = P("\\") -local single = P("'") -local double = P('"') -local left = P('{') -local right = P('}') -local both = left + right -local lineending = S("\n\r") -local space = S(" \t\n\r\f") -- / " " -local spacing = space^0 -local equal = P("=") ------ collapsed = (space^1)/ " " -local collapsed = (p_whitespace^1)/" " - ------ balanced = lpegpatterns.balanced - -local balanced = P { --- [1] = ((escape * (left+right)) + (collapsed + 1 - (left+right)) + V(2))^0, - [1] = ((escape * (left+right)) + collapsed + (1 - (left+right))^1 + V(2))^0, - [2] = left * V(1) * right, -} + local percent = P("%") + local start = P("@") + local comma = P(",") + local hash = P("#") + local escape = P("\\") + local single = P("'") + local double = P('"') + local left = P('{') + local right = P('}') + local both = left + right + local lineending = S("\n\r") + local space = S(" \t\n\r\f") -- / " " + local spacing = space^0 + local equal = P("=") + ----- collapsed = (space^1)/ " " + local collapsed = (p_whitespace^1)/" " + + ----- balanced = lpegpatterns.balanced + + local balanced = P { + -- [1] = ((escape * (left+right)) + (collapsed + 1 - (left+right)) + V(2))^0, + [1] = ((escape * (left+right)) + collapsed + (1 - (left+right))^1 + V(2))^0, + [2] = left * V(1) * right, + } -local unbalanced = P { - [1] = left * V(2) * right, - [2] = ((escape * (left+right)) + collapsed + (1 - (left+right))^1 + V(1))^0, -} + local unbalanced = P { + [1] = left * V(2) * right, + [2] = ((escape * (left+right)) + collapsed + (1 - (left+right))^1 + V(1))^0, + } -local keyword = C((R("az","AZ","09") + S("@_:-"))^1) -local key = C((1-space-equal)^1) -local tag = C((1-space-comma)^1) -local reference = keyword -local category = C((1-space-left)^1) -local s_quoted = ((escape*single) + collapsed + (1-single))^0 -local d_quoted = ((escape*double) + collapsed + (1-double))^0 + local keyword = C((R("az","AZ","09") + S("@_:-"))^1) + local key = C((1-space-equal)^1) + local tag = C((1-space-comma)^1) + local reference = keyword + local category = C((1-space-left)^1) + local s_quoted = ((escape*single) + collapsed + (1-single))^0 + local d_quoted = ((escape*double) + collapsed + (1-double))^0 -local b_value = (left /"") * balanced * (right /"") -local u_value = (left /"") * unbalanced * (right /"") -- get rid of outer { } -local s_value = (single/"") * (u_value + s_quoted) * (single/"") -local d_value = (double/"") * (u_value + d_quoted) * (double/"") -local r_value = reference * Carg(1) /resolve + local b_value = (left /"") * balanced * (right /"") + local u_value = (left /"") * unbalanced * (right /"") -- get rid of outer { } + local s_value = (single/"") * (u_value + s_quoted) * (single/"") + local d_value = (double/"") * (u_value + d_quoted) * (double/"") + local r_value = reference * Carg(1) /resolve -local somevalue = d_value + b_value + s_value + r_value -local value = Cs((somevalue * ((spacing * hash * spacing)/"" * somevalue)^0)) + local somevalue = d_value + b_value + s_value + r_value + local value = Cs((somevalue * ((spacing * hash * spacing)/"" * somevalue)^0)) -local forget = percent^1 * (1-lineending)^0 -local spacing = spacing * forget^0 * spacing -local assignment = spacing * key * spacing * equal * spacing * value * spacing -local definition = category * spacing * left * spacing * tag * spacing * comma * Ct((assignment * comma^0)^0) * spacing * right * Carg(1) / do_definition + local forget = percent^1 * (1-lineending)^0 + local spacing = spacing * forget^0 * spacing + local assignment = spacing * key * spacing * equal * spacing * value * spacing + local definition = category * spacing * left * spacing * tag * spacing * comma * Ct((assignment * comma^0)^0) * spacing * right * Carg(1) / do_definition -local crapword = C((1-space-left)^1) -local shortcut = Cmt(crapword,function(_,p,s) return lower(s) == "string" and p end) * spacing * left * ((assignment * Carg(1))/do_shortcut * comma^0)^0 * spacing * right -local comment = Cmt(crapword,function(_,p,s) return lower(s) == "comment" and p end) * spacing * lpegpatterns.argument * Carg(1) / do_comment + local crapword = C((1-space-left)^1) + local shortcut = Cmt(crapword,function(_,p,s) return lower(s) == "string" and p end) * spacing * left * ((assignment * Carg(1))/do_shortcut * comma^0)^0 * spacing * right + local comment = Cmt(crapword,function(_,p,s) return lower(s) == "comment" and p end) * spacing * lpegpatterns.argument * Carg(1) / do_comment -local casecrap = #S("sScC") * (shortcut + comment) + local casecrap = #S("sScC") * (shortcut + comment) -local bibtotable = (space + forget + P("@") * (casecrap + definition) + 1)^0 + local bibtotable = (space + forget + P("@") * (casecrap + definition) + 1)^0 --- todo \% + -- todo \% --- loadbibdata -> dataset.luadata --- loadtexdata -> dataset.luadata --- loadluadata -> dataset.luadata + -- loadbibdata -> dataset.luadata + -- loadtexdata -> dataset.luadata + -- loadluadata -> dataset.luadata --- converttoxml -> dataset.xmldata from dataset.luadata + -- converttoxml -> dataset.xmldata from dataset.luadata -function publications.loadbibdata(dataset,content,source,kind) - if not source then - report("invalid source for dataset %a",dataset) - return + function publications.loadbibdata(dataset,content,source,kind) + if not source then + report("invalid source for dataset %a",dataset) + return + end + local current = datasets[dataset] + local size = #content + if size == 0 then + report("empty source %a for dataset %a",source,current.name) + else + report("adding bib data to set %a from source %a",current.name,source) + end + statistics.starttiming(publications) + publicationsstats.nofbytes = publicationsstats.nofbytes + size + current.nofbytes = current.nofbytes + size + if source then + table.insert(current.sources, { filename = source, checksum = md5.HEX(content) }) + current.loaded[source] = kind or true + end + current.newtags = #current.luadata > 0 and { } or current.newtags + lpegmatch(bibtotable,content or "",1,current) + statistics.stoptiming(publications) end - dataset = datasets[dataset] - local size = #content - if size == 0 then - report("empty source %a for dataset %a",source,dataset.name) - else - report("adding bib data to set %a from source %a",dataset.name,source) - end - statistics.starttiming(publications) - publicationsstats.nofbytes = publicationsstats.nofbytes + size - dataset.nofbytes = dataset.nofbytes + size - if source then - table.insert(dataset.sources, { filename = source, checksum = md5.HEX(content) }) - dataset.loaded[source] = kind or true - end - dataset.newtags = #dataset.luadata > 0 and { } or dataset.newtags - publications.markasupdated(dataset) - lpegmatch(bibtotable,content or "",1,dataset) - statistics.stoptiming(publications) -end - --- we could use xmlescape again -local cleaner_0 = S('<>&') -local cleaner_1 = (1-cleaner_0)^0 * cleaner_0 -local cleaner_2 = Cs ( ( - P("<") / "<" + - P(">") / ">" + - P("&") / "&" + - P(1) -)^0) - -local compact = false -- can be a directive but then we also need to deal with newlines ... not now +end -function publications.converttoxml(dataset,nice,dontstore,usedonly) -- we have fields ! - current = datasets[dataset] - local luadata = current and current.luadata - if luadata then - statistics.starttiming(publications) - -- - local result, r, n = { }, 0, 0 - local usedonly = usedonly and publications.usedentries(dataset) - -- - r = r + 1 ; result[r] = "<?xml version='1.0' standalone='yes'?>" - r = r + 1 ; result[r] = "<bibtex>" - -- - if nice then - local f_entry_start = formatters[" <entry tag='%s' category='%s' index='%s'>"] - local s_entry_stop = " </entry>" - local f_field = formatters[" <field name='%s'>%s</field>"] - for tag, entry in sortedhash(luadata) do - if not usedonly or usedonly[tag] then - r = r + 1 ; result[r] = f_entry_start(tag,entry.category,entry.index) - for key, value in sortedhash(entry) do - if key ~= "tag" and key ~= "category" and key ~= "index" then - if lpegmatch(cleaner_1,value) then - value = lpegmatch(cleaner_2,value) - end - if value ~= "" then - r = r + 1 ; result[r] = f_field(key,value) +do + + -- we could use xmlescape again + + local cleaner_0 = S('<>&') + local cleaner_1 = (1-cleaner_0)^0 * cleaner_0 + local cleaner_2 = Cs ( ( + P("<") / "<" + + P(">") / ">" + + P("&") / "&" + + P(1) + )^0) + + local compact = false -- can be a directive but then we also need to deal with newlines ... not now + + function publications.converttoxml(dataset,nice,dontstore,usedonly) -- we have fields ! + local current = datasets[dataset] + local luadata = current and current.luadata + if luadata then + statistics.starttiming(publications) + -- + local result, r, n = { }, 0, 0 + local usedonly = usedonly and publications.usedentries() + -- + r = r + 1 ; result[r] = "<?xml version='1.0' standalone='yes'?>" + r = r + 1 ; result[r] = "<bibtex>" + -- + if nice then + local f_entry_start = formatters[" <entry tag='%s' category='%s' index='%s'>"] + local s_entry_stop = " </entry>" + local f_field = formatters[" <field name='%s'>%s</field>"] + for tag, entry in sortedhash(luadata) do + if not usedonly or usedonly[tag] then + r = r + 1 ; result[r] = f_entry_start(tag,entry.category,entry.index) + for key, value in sortedhash(entry) do + if key ~= "tag" and key ~= "category" and key ~= "index" then + if lpegmatch(cleaner_1,value) then + value = lpegmatch(cleaner_2,value) + end + if value ~= "" then + r = r + 1 ; result[r] = f_field(key,value) + end end end + r = r + 1 ; result[r] = s_entry_stop + n = n + 1 end - r = r + 1 ; result[r] = s_entry_stop - n = n + 1 end - end - else - local f_entry_start = formatters["<entry tag='%s' category='%s' index='%s'>"] - local s_entry_stop = "</entry>" - local f_field = formatters["<field name='%s'>%s</field>"] - for tag, entry in next, luadata do - if not usedonly or usedonly[tag] then - r = r + 1 ; result[r] = f_entry_start(entry.tag,entry.category,entry.index) - for key, value in next, entry do - if key ~= "tag" and key ~= "category" and key ~= "index" then - if lpegmatch(cleaner_1,value) then - value = lpegmatch(cleaner_2,value) - end - if value ~= "" then - r = r + 1 ; result[r] = f_field(key,value) + else + local f_entry_start = formatters["<entry tag='%s' category='%s' index='%s'>"] + local s_entry_stop = "</entry>" + local f_field = formatters["<field name='%s'>%s</field>"] + for tag, entry in next, luadata do + if not usedonly or usedonly[tag] then + r = r + 1 ; result[r] = f_entry_start(entry.tag,entry.category,entry.index) + for key, value in next, entry do + if key ~= "tag" and key ~= "category" and key ~= "index" then + if lpegmatch(cleaner_1,value) then + value = lpegmatch(cleaner_2,value) + end + if value ~= "" then + r = r + 1 ; result[r] = f_field(key,value) + end end end + r = r + 1 ; result[r] = s_entry_stop + n = n + 1 end - r = r + 1 ; result[r] = s_entry_stop - n = n + 1 end end - end - -- - r = r + 1 ; result[r] = "</bibtex>" - -- - result = concat(result,nice and "\n" or nil) - -- - if dontstore then - -- indeed - else - statistics.starttiming(xml) - current.xmldata = xmlconvert(result, { - resolve_entities = true, - resolve_predefined_entities = true, -- in case we have escaped entities - -- unify_predefined_entities = true, -- & -> & - utfize_entities = true, - } ) - statistics.stoptiming(xml) - if lxml then - lxml.register(formatters["btx:%s"](current.name),current.xmldata) + -- + r = r + 1 ; result[r] = "</bibtex>" + -- + result = concat(result,nice and "\n" or nil) + -- + if dontstore then + -- indeed + else + statistics.starttiming(xml) + current.xmldata = xmlconvert(result, { + resolve_entities = true, + resolve_predefined_entities = true, -- in case we have escaped entities + -- unify_predefined_entities = true, -- & -> & + utfize_entities = true, + } ) + statistics.stoptiming(xml) + if lxml then + lxml.register(formatters["btx:%s"](current.name),current.xmldata) + end end + statistics.stoptiming(publications) + return result, n end - statistics.stoptiming(publications) - return result, n end + end -local loaders = publications.loaders or { } -publications.loaders = loaders +do -local function resolvedname(dataset,filename) - dataset = datasets[dataset] - if type(filename) ~= "string" then - report("invalid filename %a",tostring(filename)) - end - local fullname = resolvers.findfile(filename,"bib") - if fullname == "" then - fullname = resolvers.findfile(filename) -- let's not be too picky - end - if not fullname or fullname == "" then - report("no file %a",filename) - dataset.fullname = filename - return dataset, false - else - dataset.fullname = fullname - return dataset, fullname + local function resolvedname(dataset,filename) + local current = datasets[dataset] + if type(filename) ~= "string" then + report("invalid filename %a",tostring(filename)) + end + local fullname = resolvers.findfile(filename,"bib") + if fullname == "" then + fullname = resolvers.findfile(filename) -- let's not be too picky + end + if not fullname or fullname == "" then + report("no file %a",filename) + current.fullname = filename + return current, false + else + current.fullname = fullname + return current, fullname + end end -end -publications.resolvedname = resolvedname + publications.resolvedname = resolvedname -function loaders.bib(dataset,filename,kind) - local dataset, fullname = resolvedname(dataset,filename) - if not fullname then - return + function loaders.bib(dataset,filename,kind) + local dataset, fullname = resolvedname(dataset,filename) + if not fullname then + return + end + local data = io.loaddata(filename) or "" + if data == "" then + report("empty file %a, nothing loaded",fullname) + return + end + if trace then + report("loading file",fullname) + end + publications.loadbibdata(dataset,data,fullname,kind) end - local data = io.loaddata(filename) or "" - if data == "" then - report("empty file %a, nothing loaded",fullname) - return + + function loaders.lua(dataset,filename) -- if filename is a table we load that one + local current, data, fullname + if type(filename) == "table" then + current = datasets[dataset] + data = filename + else + dataset, fullname = resolvedname(dataset,filename) + if not fullname then + return + end + current = datasets[dataset] + data = table.load(filename) + end + if data then + local luadata = current.luadata + for tag, entry in next, data do + if type(entry) == "table" then + entry.index = getindex(current,luadata,tag) + entry.tag = tag + luadata[tag] = entry -- no cleaning yet + end + end + end end - if trace then - report("loading file",fullname) + + function loaders.buffer(dataset,name) -- if filename is a table we load that one + local current = datasets[dataset] + local barename = file.removesuffix(name) + local data = buffers.getcontent(barename) or "" + if data == "" then + report("empty buffer %a, nothing loaded",barename) + return + end + if trace then + report("loading buffer",barename) + end + publications.loadbibdata(current,data,barename,"bib") end - publications.loadbibdata(dataset,data,fullname,kind) -end -function loaders.lua(dataset,filename) -- if filename is a table we load that one - local data, fullname - if type(filename) == "table" then - dataset = datasets[dataset] - data = filename - else - dataset, fullname = resolvedname(dataset,filename) + function loaders.xml(dataset,filename) + local dataset, fullname = resolvedname(dataset,filename) if not fullname then return end - data = table.load(filename) - end - if data then - local luadata = dataset.luadata - for tag, entry in next, data do - if type(entry) == "table" then - entry.index = getindex(dataset,luadata,tag) - entry.tag = tag - luadata[tag] = entry -- no cleaning yet + local current = datasets[dataset] + local luadata = current.luadata + local root = xml.load(filename) + for bibentry in xmlcollected(root,"/bibtex/entry") do + local attributes = bibentry.at + local tag = attributes.tag + local entry = { + category = attributes.category, + tag = tag, -- afterwards also set, to prevent overload + index = 0, -- prelocated + } + for field in xmlcollected(bibentry,"/field") do + entry[field.at.name] = field.dt[1] -- no cleaning yet | xmltext(field) end + entry.index = getindex(current,luadata,tag) + entry.tag = tag + luadata[tag] = entry end end -end -function loaders.buffer(dataset,name) -- if filename is a table we load that one - dataset = datasets[dataset] - name = file.removesuffix(name) - local data = buffers.getcontent(name) or "" - if data == "" then - report("empty buffer %a, nothing loaded",name) - return - end - if trace then - report("loading buffer",name) - end - publications.loadbibdata(dataset,data,name,"bib") -end + setmetatableindex(loaders,function(t,filetype) + local v = function(dataset,filename) + report("no loader for file %a with filetype %a",filename,filetype) + end + t[k] = v + return v + end) -function loaders.xml(dataset,filename) - local dataset, fullname = resolvedname(dataset,filename) - if not fullname then - return - end - local luadata = dataset.luadata - local root = xml.load(filename) - for bibentry in xmlcollected(root,"/bibtex/entry") do - local attributes = bibentry.at - local tag = attributes.tag - local entry = { - category = attributes.category, - tag = tag, -- afterwards also set, to prevent overload - index = 0, -- prelocated - } - for field in xmlcollected(bibentry,"/field") do - entry[field.at.name] = field.dt[1] -- no cleaning yet | xmltext(field) + function publications.load(specification) + local current = datasets[specification.dataset or v_standard] + local files = settings_to_array(specification.filename) + local kind = specification.kind + local dataspec = specification.specification + statistics.starttiming(publications) + for i=1,#files do + local filetype, filename = string.splitup(files[i],"::") + if not filename then + filename = filetype + filetype = file.suffix(filename) + end + if filename then + if not filetype or filetype == "" then + filetype = "bib" + end + if file.suffix(filename) == "" then + file.addsuffix(filename,filetype) + end + loaders[filetype](current,filename) + if kind then + current.loaded[current.fullname or filename] = kind + end + if dataspec then + current.specifications[dataspec] = true + end + end + end + local runner = enhancer.runner + if runner then + runner(current) end - entry.index = getindex(dataset,luadata,tag) - entry.tag = tag - luadata[tag] = entry + statistics.stoptiming(publications) + return current end + end -setmetatableindex(loaders,function(t,filetype) - local v = function(dataset,filename) - report("no loader for file %a with filetype %a",filename,filetype) - end - t[k] = v - return v -end) +do -function publications.load(dataset,filename,kind) - dataset = datasets[dataset] - statistics.starttiming(publications) - local files = settings_to_array(filename) - for i=1,#files do - local filetype, filename = string.splitup(files[i],"::") - if not filename then - filename = filetype - filetype = file.suffix(filename) + function enhancers.order(dataset) + local luadata = dataset.luadata + local ordered = sortedkeys(luadata) + local total = #ordered + for i=1,total do + ordered[i] = luadata[ordered[i]] end - if filename then - if not filetype or filetype == "" then - filetype = "bib" - end - if file.suffix(filename) == "" then - file.addsuffix(filename,filetype) - end - loaders[filetype](dataset,filename) - if kind then - dataset.loaded[dataset.fullname or filename] = kind + dataset.ordered = ordered + end + + function enhancers.details(dataset) + local luadata = dataset.luadata + local details = dataset.details + for tag, entry in next, luadata do + if not details[tag] then + details[tag] = { } end end end - statistics.stoptiming(publications) - return dataset + + utilities.sequencers.appendaction(enhancer,"system","publications.enhancers.order") + utilities.sequencers.appendaction(enhancer,"system","publications.enhancers.details") + end -local checked = function(s,d) d[s] = (d[s] or 0) + 1 end -local checktex = ( (1-P("\\"))^1 + P("\\") * ((C(R("az","AZ")^1) * Carg(1))/checked))^0 - -function publications.analyze(dataset) - dataset = datasets[dataset] - local data = dataset.luadata - local categories = { } - local fields = { } - local commands = { } - for k, v in next, data do - categories[v.category] = (categories[v.category] or 0) + 1 - for k, v in next, v do - fields[k] = (fields[k] or 0) + 1 - lpegmatch(checktex,v,1,commands) +do + + local checked = function(s,d) d[s] = (d[s] or 0) + 1 end + local checktex = ( (1-P("\\"))^1 + P("\\") * ((C(R("az","AZ")^1) * Carg(1))/checked))^0 + + function publications.analyze(dataset) + local current = datasets[dataset] + local data = current.luadata + local categories = { } + local fields = { } + local commands = { } + for k, v in next, data do + categories[v.category] = (categories[v.category] or 0) + 1 + for k, v in next, v do + fields[k] = (fields[k] or 0) + 1 + lpegmatch(checktex,v,1,commands) + end end + current.analysis = { + categories = categories, + fields = fields, + commands = commands, + } end - dataset.analysis = { - categories = categories, - fields = fields, - commands = commands, - } -end --- str = [[ --- @COMMENT { CRAP } --- @STRING{ hans = "h a n s" } --- @STRING{ taco = "t a c o" } --- @SOMETHING{ key1, abc = "t a c o" , def = "h a n s" } --- @SOMETHING{ key2, abc = hans # taco } --- @SOMETHING{ key3, abc = "hans" # taco } --- @SOMETHING{ key4, abc = hans # "taco" } --- @SOMETHING{ key5, abc = hans # taco # "hans" # "taco"} --- @SOMETHING{ key6, abc = {oeps {oeps} oeps} } --- ]] - --- local dataset = publications.new() --- publications.tolua(dataset,str) --- publications.toxml(dataset) --- publications.toxml(dataset) --- print(dataset.xmldata) --- inspect(dataset.luadata) --- inspect(dataset.xmldata) --- inspect(dataset.shortcuts) --- print(dataset.nofbytes,statistics.elapsedtime(publications)) - --- local dataset = publications.new() --- publications.load(dataset,"IEEEabrv.bib") --- publications.load(dataset,"IEEEfull.bib") --- publications.load(dataset,"IEEEexample.bib") --- publications.toxml(dataset) --- print(dataset.nofbytes,statistics.elapsedtime(publications)) - --- local dataset = publications.new() --- publications.load(dataset,"gut.bib") --- publications.load(dataset,"komoedie.bib") --- publications.load(dataset,"texbook1.bib") --- publications.load(dataset,"texbook2.bib") --- publications.load(dataset,"texbook3.bib") --- publications.load(dataset,"texgraph.bib") --- publications.load(dataset,"texjourn.bib") --- publications.load(dataset,"texnique.bib") --- publications.load(dataset,"tugboat.bib") --- publications.toxml(dataset) --- print(dataset.nofbytes,statistics.elapsedtime(publications)) - --- print(table.serialize(dataset.luadata)) --- print(table.serialize(dataset.xmldata)) --- print(table.serialize(dataset.shortcuts)) --- print(xml.serialize(dataset.xmldata)) +end -- a helper: @@ -915,91 +918,115 @@ end -- savers -local savers = { } +do -local s_preamble = [[ -% this is an export from context mkiv + local savers = { } -@preamble{ - \ifdefined\btxcmd - % we're probably in context - \else - \def\btxcmd#1{\csname#1\endcsname} - \fi -} + local s_preamble = [[ + % this is an export from context mkiv -]] - -function savers.bib(dataset,filename,usedonly) - local current = datasets[dataset] - local luadata = current.luadata or { } - local usedonly = usedonly and publications.usedentries(dataset) - local f_start = formatters["@%s{%s,\n"] - local f_field = formatters[" %s = {%s},\n"] - local s_stop = "}\n\n" - local result = { s_preamble } - local n, r = 0, 1 - for tag, data in sortedhash(luadata) do - if not usedonly or usedonly[tag] then - r = r + 1 ; result[r] = f_start(data.category or "article",tag) - for key, value in sortedhash(data) do - if privates[key] then - -- skip - else - r = r + 1 ; result[r] = f_field(key,value) + @preamble{ + \ifdefined\btxcmd + % we're probably in context + \else + \def\btxcmd#1{\csname#1\endcsname} + \fi + } + + ]] + + function savers.bib(dataset,filename,usedonly) + local current = datasets[dataset] + local luadata = current.luadata or { } + local usedonly = usedonly and publications.usedentries() + local f_start = formatters["@%s{%s,\n"] + local f_field = formatters[" %s = {%s},\n"] + local s_stop = "}\n\n" + local result = { s_preamble } + local n, r = 0, 1 + for tag, data in sortedhash(luadata) do + if not usedonly or usedonly[tag] then + r = r + 1 ; result[r] = f_start(data.category or "article",tag) + for key, value in sortedhash(data) do + if privates[key] then + -- skip + else + r = r + 1 ; result[r] = f_field(key,value) + end end + r = r + 1 ; result[r] = s_stop + n = n + 1 end - r = r + 1 ; result[r] = s_stop - n = n + 1 end + report("%s entries from dataset %a saved in %a",n,dataset,filename) + io.savedata(filename,concat(result)) end - report("%s entries from dataset %a saved in %a",n,dataset,filename) - io.savedata(filename,concat(result)) -end -function savers.lua(dataset,filename,usedonly) - local current = datasets[dataset] - local luadata = current.luadata or { } - local usedonly = usedonly and publications.usedentries(dataset) - if usedonly then - local list = { } - for k, v in next, luadata do - if usedonly[k] then - list[k] = v + function savers.lua(dataset,filename,usedonly) + local current = datasets[dataset] + local luadata = current.luadata or { } + local usedonly = usedonly and publications.usedentries() + if usedonly then + local list = { } + for k, v in next, luadata do + if usedonly[k] then + list[k] = v + end end + luadata = list end - luadata = list + report("%s entries from dataset %a saved in %a",table.count(luadata),dataset,filename) + table.save(filename,luadata) end - report("%s entries from dataset %a saved in %a",table.count(luadata),dataset,filename) - table.save(filename,luadata) -end -function savers.xml(dataset,filename,usedonly) - local result, n = publications.converttoxml(dataset,true,true,usedonly) - report("%s entries from dataset %a saved in %a",n,dataset,filename) - io.savedata(filename,result) -end - -function publications.save(dataset,filename,kind,usedonly) - statistics.starttiming(publications) - if not kind or kind == "" then - kind = file.suffix(filename) + function savers.xml(dataset,filename,usedonly) + local result, n = publications.converttoxml(dataset,true,true,usedonly) + report("%s entries from dataset %a saved in %a",n,dataset,filename) + io.savedata(filename,result) end - local saver = savers[kind] - if saver then - usedonly = usedonly ~= v_all - saver(dataset,filename,usedonly) - else - report("unknown format %a for saving %a",kind,dataset) + + function publications.save(dataset,filename,kind,usedonly) + statistics.starttiming(publications) + if not kind or kind == "" then + kind = file.suffix(filename) + end + local saver = savers[kind] + if saver then + usedonly = usedonly ~= v_all + saver(dataset,filename,usedonly) + else + report("unknown format %a for saving %a",kind,dataset) + end + statistics.stoptiming(publications) + return dataset end - statistics.stoptiming(publications) - return dataset + + commands.btxsavedataset = publications.save + end -commands.btxsavedataset = publications.save +-- casters + +do --- loaders.bib("test",resolvers.findfile("mkiv-publications.bib","bibtex")) --- --- publications.save("test","e:/tmp/foo.bib") --- publications.save("test","e:/tmp/foo.lua") --- publications.save("test","e:/tmp/foo.xml") + publications.detailed = setmetatableindex(function(detailed,kind) + local values = setmetatableindex(function(values,value) + local caster = casters[kind] + local cast = caster and caster(value) or value + values[value] = cast + return cast + end) + detailed[kind] = values + return values + end) + + casters.keyword = utilities.parsers.settings_to_set + + local pagessplitter = lpeg.splitat(P("-")^1) + + casters.pagenumber = function(str) + local first, last = lpegmatch(pagessplitter,str) + return first and last and { first, last } or str + end + +end diff --git a/tex/context/base/publ-fnd.lua b/tex/context/base/publ-fnd.lua index 739ca9d29..5308302cb 100644 --- a/tex/context/base/publ-fnd.lua +++ b/tex/context/base/publ-fnd.lua @@ -13,11 +13,9 @@ end -- this tracker is only for real debugging and not for the average user -local trace_match = false trackers.register("publications.cite.match", function(v) trace_match = v end) +local trace_match = false trackers.register("publications.match", function(v) trace_match = v end) -if not publications then - publications = { } -end +local publications = publications local tonumber, next, type = tonumber, next, type local find = string.find @@ -29,7 +27,7 @@ local concat = table.concat local formatters = string.formatters local lowercase = characters.lower -local report = logs.reporter("publications","match") +local report = logs.reporter("publications","match") local colon = P(":") local dash = P("-") diff --git a/tex/context/base/publ-imp-apa.lua b/tex/context/base/publ-imp-apa.lua index 3ee67f6d1..9b68fdc9b 100644 --- a/tex/context/base/publ-imp-apa.lua +++ b/tex/context/base/publ-imp-apa.lua @@ -181,7 +181,7 @@ categories.inbook = { "volume", "number", "series", "edition", "month", "address", - "note", "isbn", + "note", "isbn" }, } @@ -199,7 +199,7 @@ categories.booklet = { "subtitle", "type", "file", "address", "howpublished", - "note", "isbn", + "note", "isbn" }, } @@ -278,8 +278,22 @@ categories.thesis = { }, } -categories.mastersthesis = categories.thesis -categories.phdthesis = categories.thesis +categories.mastersthesis = { + required = { + "author", + "title", + "school", + "year" + }, + optional = { + "type", + "subtitle", "file", + "month", + "address", + "note" + }, +} +categories.phdthesis = categories.mastersthesis -- a report published by a school or other institution, usually numbered within a series. diff --git a/tex/context/base/publ-ini.lua b/tex/context/base/publ-ini.lua index 188995b0a..b7f164610 100644 --- a/tex/context/base/publ-ini.lua +++ b/tex/context/base/publ-ini.lua @@ -6,6 +6,9 @@ if not modules then modules = { } end modules ['publ-ini'] = { license = "see context related readme files" } +-- bah .. this 200 locals limit again ... so we need to split it as adding more +-- do ... ends makes it messier + -- plug the list sorted in the list mechanism (specification.sortorder) -- todo: delay details till alternative is known so that potential author @@ -18,7 +21,7 @@ if not modules then modules = { } end modules ['publ-ini'] = { -- load big bib files many times and even then ... fonts are larger. local next, rawget, type, tostring, tonumber = next, rawget, type, tostring, tonumber -local match, gmatch, format, gsub, find = string.match, string.gmatch, string.format, string.gsub, string.find +local match, find = string.match, string.find local concat, sort, tohash = table.concat, table.sort, table.tohash local utfsub = utf.sub local mod = math.mod @@ -38,10 +41,15 @@ local trace = false trackers.register("publications", local trace_cite = false trackers.register("publications.cite", function(v) trace_cite = v end) local trace_missing = false trackers.register("publications.cite.missing", function(v) trace_missing = v end) local trace_references = false trackers.register("publications.cite.references", function(v) trace_references = v end) +local trace_detail = false trackers.register("publications.detail", function(v) trace_detail = v end) publications = publications or { } local datasets = publications.datasets local writers = publications.writers +local casters = publications.casters +local detailed = publications.detailed +local enhancer = publications.enhancer +local enhancers = publications.enhancers local tracers = publications.tracers or { } publications.tracers = tracers @@ -162,17 +170,12 @@ function commands.registerbtxlistvariant(name,parent) registeredlistvariants[name] = parent or "" end -local specifications = publications.specifications -local currentspecification = specifications[false] ------ currentspecificationfields = currentspecification.fields -local currentspecificationcategories = currentspecification.categories - -local ignoredfields = { } +local specifications = publications.specifications +local currentspecification = specifications[false] +local ignoredfields = { } local function setspecification(name) - currentspecification = specifications[name] - -- currentspecificationfields = currentspecification.fields - currentspecificationcategories = currentspecification.categories + currentspecification = specifications[name] if trace then report("setting specification %a",type(name) == "string" and name or "anything") end @@ -238,9 +241,6 @@ end) local collected = allocate() local tobesaved = allocate() --- we use a a dedicated (and efficient as it know what it deals with) serializer, --- also because we need to ignore the 'details' field - do local function serialize(t) @@ -312,7 +312,11 @@ do if datasources then for i=1,#datasources do local filename = datasources[i].filename - publications.load(dataset,filename,"previous") + publications.load { + dataset = dataset, + filename = filename, + kind = "previous" + } end end if usersource then @@ -327,21 +331,19 @@ do end -if not publications.authors then - initializer() -- for now, runtime loaded -end - -- we want to minimize references as there can be many (at least -- when testing) -local initialized = false -local usedentries = allocate { } -local citetolist = allocate { } -local listtocite = allocate { } local nofcitations = 0 +local usedentries = nil +local citetolist = nil +local listtocite = nil -setmetatableindex(usedentries,function(t,k) - if not initialized then +do + + local initialize = nil + + initialize = function(t) usedentries = allocate { } citetolist = allocate { } listtocite = allocate { } @@ -434,12 +436,20 @@ setmetatableindex(usedentries,function(t,k) end end end - return usedentries[k] + initialize = nil + end + + usedentries = setmetatableindex(function(_,k) if initialize then initialize() end return usedentries[k] end) + citetolist = setmetatableindex(function(_,k) if initialize then initialize() end return citetolist [k] end) + listtocite = setmetatableindex(function(_,k) if initialize then initialize() end return listtocite [k] end) + + function publications.usedentries() + if initialize then + initialize() + end + return usedentries end -end) -function publications.usedentries(dataset) - return usedentries[dataset] end -- match: @@ -451,103 +461,108 @@ end -- by prefix -- by dataset -local reported = { } -local finder = publications.finder - -local function findallused(dataset,reference,internal) - local finder = publications.finder -- for the moment, not yet in all betas - local find = finder and finder(reference) - local tags = not find and settings_to_array(reference) - local todo = { } - local okay = { } -- only if mark - local set = usedentries[dataset] - local valid = datasets[dataset].luadata - local ordered = datasets[dataset].ordered - if set then - local function register(tag) - local entry = set[tag] - if entry then - -- only once in a list but at some point we can have more (if we - -- decide to duplicate) - if #entry == 1 then - entry = entry[1] - else - -- same block and section - local done = false - if internal and internal > 0 then - -- first following in list - for i=1,#entry do - local e = entry[i] - if e.references.internal > internal then - done = e - break - end - end - if not done then - -- last preceding in list +local findallused do + + local reported = { } + local finder = publications.finder + + findallused = function(dataset,reference,internal) + local finder = publications.finder -- for the moment, not yet in all betas + local find = finder and finder(reference) + local tags = not find and settings_to_array(reference) + local todo = { } + local okay = { } -- only if mark + local set = usedentries[dataset] + local current = datasets[dataset] + local valid = current.luadata + local ordered = current.ordered + if set then + local function register(tag) + local entry = set[tag] + if entry then + -- only once in a list but at some point we can have more (if we + -- decide to duplicate) + if #entry == 1 then + entry = entry[1] + else + -- same block and section + local done = false + if internal and internal > 0 then + -- first following in list for i=1,#entry do local e = entry[i] - if e.references.internal < internal then + if e.references.internal > internal then done = e - else break end end + if not done then + -- last preceding in list + for i=1,#entry do + local e = entry[i] + if e.references.internal < internal then + done = e + else + break + end + end + end + end + if done then + entry = done + else + entry = entry[1] end end - if done then - entry = done - else - entry = entry[1] - end - end - okay[#okay+1] = entry - end - todo[tag] = true - end - if find then - tags = { } - for i=1,#ordered do - local entry = ordered[i] - if find(entry) then - local tag = entry.tag - register(tag) - tags[#tags+1] = tag + okay[#okay+1] = entry end + todo[tag] = true end - else - for i=1,#tags do - local tag = tags[i] - if valid[tag] then - register(tag) - elseif not reported[tag] then - reported[tag] = true - report_cite("non-existent entry %a in %a",tag,dataset) + if find then + tags = { } + for i=1,#ordered do + local entry = ordered[i] + if find(entry) then + local tag = entry.tag + register(tag) + tags[#tags+1] = tag + end end - end - end - else - if find then - tags = { } - for i=1,#ordered do - local entry = ordered[i] - if find(entry) then - tags[#tags+1] = entry.tag + else + for i=1,#tags do + local tag = tags[i] + if valid[tag] then + register(tag) + elseif not reported[tag] then + reported[tag] = true + report_cite("non-existent entry %a in %a",tag,dataset) + end end end else - for i=1,#tags do - local tag = tags[i] - if valid[tag] then - todo[tag] = true - elseif not reported[tag] then - reported[tag] = true - report_cite("non-existent entry %a in %a",tag,dataset) + if find then + tags = { } + for i=1,#ordered do + local entry = ordered[i] + if find(entry) then + tags[#tags+1] = entry.tag + end + end + else + for i=1,#tags do + local tag = tags[i] + if valid[tag] then + todo[tag] = true + elseif not reported[tag] then + reported[tag] = true + report_cite("non-existent entry %a in %a",tag,dataset) + end end end end + return okay, todo, tags end - return okay, todo, tags + end local function unknowncite(reference) @@ -671,201 +686,132 @@ end -- basic loading -function commands.usebtxdataset(name,filename) - publications.load(datasets[name],filename,"current") +function commands.usebtxdataset(specification) + specification.kind = "current" + publications.load(specification) end function commands.convertbtxdatasettoxml(name,nice) - publications.converttoxml(datasets[name],nice) + publications.converttoxml(name,nice) end -- enhancing -local splitauthorstring = publications.authors.splitstring - -local pagessplitter = lpeg.splitat(P("-")^1) - --- maybe not redo when already done +do -local function shortsorter(a,b) - local ay, by = a[2], b[2] - if ay ~= by then - return ay < by - end - local ay, by = a[3], b[3] - if ay ~= by then - return ay < by - end - return a[4] < b[4] -end + -- maybe not redo when already done -function publications.enhance(dataset) -- for the moment split runs (maybe publications.enhancers) - statistics.starttiming(publications) - if type(dataset) == "string" then - dataset = datasets[dataset] - else - -- assume table, otherwise maybe issue an error - end - local used = usedentries[dataset.name] or { } - local luadata = dataset.luadata - local details = dataset.details - local ordered = dataset.ordered - local types = currentspecification.types -- or { author = "author", editor = "author" } - local authors = { } - -- - for k, v in next, types do - if v == "author" then - authors[#authors+1] = k + local function shortsorter(a,b) + local ay, by = a[2], b[2] + if ay ~= by then + return ay < by end - end - -- - for tag, entry in next, luadata do - local detail = { } - details[tag] = detail - for i=1,#authors do - local key = authors[i] - local value = entry[key] - if value then - detail[key] = splitauthorstring(value) - end + local ay, by = a[3], b[3] + if ay ~= by then + return ay < by end + return a[4] < b[4] end - -- short - local shorts = { } - for i=1,#ordered do - local entry = ordered[i] - if entry then - local tag = entry.tag - if tag then - local detail = details[tag] - if detail then - local author = detail.author - if author then - -- number depends on sort order - local t = { } - if #author == 0 then - -- what - else - local n = #author == 1 and 3 or 1 - for i=1,#author do - local surnames = author[i].surnames - if not surnames or #surnames == 0 then - -- error + + -- We could avoid loops by combining enhancers but that makes it only + -- more messy and for documents that use publications the few extra milli + -- seconds are irrelevant (there is for sure more to gain by proper coding + -- of the source and or style). + + function publications.enhancers.suffixes(dataset) + local used = usedentries[dataset.name] + local luadata = dataset.luadata + local details = dataset.details + local ordered = dataset.ordered + local caster = casters.author + local getter = publications.directget + local shorts = { } + for i=1,#ordered do + local entry = ordered[i] + if entry then + local tag = entry.tag + if tag then + local use = used[tag] + if use then + -- use is a table of used list entries (so there can be more) and we just look at + -- the first one for btx properties + local listentry = use[1] + local userdata = listentry.userdata + local btxspc = userdata and userdata.btxspc + if btxspc then + local author = getter(dataset,entry,"author",specifications[btxspc]) + if author then + author = caster(author) + -- number depends on sort order + local t = { } + if #author > 0 then + local n = #author == 1 and 3 or 1 + for i=1,#author do + local surnames = author[i].surnames + if not surnames or #surnames == 0 then + -- error + else + t[#t+1] = utfsub(surnames[1],1,n) + end + end + end + local year = tonumber(entry.year) or 0 + local short = formatters["%t%02i"](t,mod(year,100)) + local s = shorts[short] + -- we could also sort on reference i.e. entries.text + if u then + u = listentry.entries.text -- hm + else + u = "0" + end + if not s then + shorts[short] = { { tag, year, u, i } } else - t[#t+1] = utfsub(surnames[1],1,n) + s[#s+1] = { tag, year, u, i } end end - end - local year = tonumber(entry.year) or 0 - local short = formatters["%t%02i"](t,mod(year,100)) - local s = shorts[short] - -- we could also sort on reference i.e. entries.text - local u = used[tag] - if u then - u = u[1].entries.text -- hm else - u = "0" + --- no spec so let's forget about it end - if not s then - shorts[short] = { { tag, year, u, i } } - else - s[#s+1] = { tag, year, u, i } - end - else - -- end - else - report("internal error, no detail for tag %s",tag) - end - -- - local pages = entry.pages or entry.page - if pages then - local first, last = lpegmatch(pagessplitter,pages) - detail.pages = first and last and { first, last } or pages - end - -- - local keyword = entry.keyword - if keyword then - detail.keyword = settings_to_set(keyword) end - -- - if category == "inbook" then - detail.maintitle = entry.chapter or entry.title - elseif category == "incollection" then - detail.maintitle = entry.title or entry.booktitle - else - detail.maintitle = entry.title or entry.chapter or entry.booktitle - end - else - report("internal error, no tag at index %s",i) end - else - report("internal error, no entry at index %s",i) - end - end - for short, tags in next, shorts do -- ordered ? - local done = #tags > 0 - -- we only use suffixes when there are multiple references to same years - -- so we check for used first - if done then - local n = 0 - for i=1,#tags do - local tag = tags[i][1] - if used[tag] then - n = n + 1 - if n > 1 then - break + end + for short, tags in next, shorts do -- ordered ? + local done = #tags > 0 + -- now we assign the suffixes, unless we have only one reference + if done then + sort(tags,shortsorter) + local n = #tags + if n > 1 then + for i=1,n do + local tag = tags[i][1] + local detail = details[tag] + local suffix = numbertochar(i) + local entry = luadata[tag] + local year = entry.year + detail.short = short + detail.suffix = suffix + if year then + detail.suffixedyear = year .. suffix + end end end - end - done = n > 1 - end - -- now we assign the suffixes, unless we have only one reference - if done then - sort(tags,shortsorter) - local n = 0 - for i=1,#tags do - local tag = tags[i][1] + else + local tag = tags[1][1] local detail = details[tag] + local entry = luadata[tag] + local year = entry.year detail.short = short - if used[tag] then - n = n + 1 - local suffix = numbertochar(n) - detail.suffix = suffix - local entry = luadata[tag] - local year = entry.year - if year then - detail.suffixedyear = year .. suffix - end + if year then + detail.suffixedyear = year end end - -- for i=1,#tags do - -- local tag = tags[i][1] - -- local detail = details[tag] - -- if not detail.suffix then - -- n = n + 1 - -- local suffix = numbertochar(n) - -- detail.suffix = suffix - -- local entry = luadata[tag] - -- local year = entry.year - -- if year then - -- detail.suffixedyear = year .. suffix - -- end - -- end - -- end - else - local tag = tags[1][1] - local detail = details[tag] - detail.short = short - local entry = luadata[tag] - local year = entry.year - if year then - detail.suffixedyear = year - end end end - dataset.enhanced = true - statistics.stoptiming(publications) + + utilities.sequencers.appendaction(enhancer,"system","publications.enhancers.suffixes") + end function commands.addbtxentry(name,settings,content) @@ -940,23 +886,28 @@ do -- when we have a special kind of database local function permitted(category,field) - local catspec = currentspecificationcategories[category] + local catspec = currentspecification.categories[category] if not catspec then - report("invalid category %a, %s",category,"no specification") + report("invalid category %a, %s",category,"no specification") -- can't happen return false end local fields = catspec.fields if not fields then - report("invalid category %a, %s",category,"no fields") + report("invalid category %a, %s",category,"no fields") -- can't happen return false end - local kind = fields[field] if ignoredfields and ignoredfields[field] then return false - else - --- just "return true" as it's already in fields - local sets = catspec.sets - return sets and sets[field] or true + end + local sets = catspec.sets + if sets then + local set = sets[field] + if set then + return set + end + end + if fields[field] then + return true end end @@ -968,36 +919,36 @@ do return field, okay end local details = dataset.details[tag] - local okay = details[field] - if okay then - return field, okay + local value = details[field] + if value then + return field, value end elseif valid then -- local fields = dataset.luadata[tag] for i=1,#valid do local field = valid[i] - local okay = fields[field] - if okay then - return field, okay + local value = fields[field] + if value then + return field, value end end local details = dataset.details[tag] for i=1,#valid do - local okay = details[field] - if okay then - return field, okay + local value = details[field] + if value then + return field, value end end end end - local function get(name,tag,field,what,check) - local dataset = rawget(datasets,name) + local function get(dataset,tag,field,what,check,catspec) + dataset = rawget(datasets,name) if dataset then local data = dataset.luadata[tag] if data then local category = data.category - local catspec = currentspecificationcategories[category] + local catspec = (catspec or currentspecification).categories[category] if not catspec then return false end @@ -1036,6 +987,32 @@ do return "" end + publications.get = get + + function publications.directget(dataset,data,field,catspec) + local catspec = (catspec or currentspecification).categories[data.category] + if not catspec then + return false + end + local fields = catspec.fields + if fields then + local sets = catspec.sets + if sets then + local set = sets[field] + if set then + for i=1,#set do + local field = set[i] + local value = fields[field] and data[field] -- redundant check + if value then + return value + end + end + end + end + return fields[field] and data[field] or nil -- redundant check + end + end + function commands.btxfieldname(name,tag,field) context(get(name,tag,field,false,false)) end function commands.btxfieldtype(name,tag,field) context(get(name,tag,field,true, false)) end function commands.btxfoundname(name,tag,field) context(get(name,tag,field,false,true )) end @@ -1053,10 +1030,10 @@ do local name, value = found(dataset,tag,field,valid,fields) if value then typesetters[currentspecification.types[name]](field,value,manipulator) - else + elseif trace_detail then report("%s %s %a in category %a for tag %a in dataset %a","unknown","entry",field,category,tag,name) end - else + elseif trace_detail then report("%s %s %a in category %a for tag %a in dataset %a","invalid","entry",field,category,tag,name) end else @@ -1078,10 +1055,10 @@ do local value = fields[field] if value then typesetters[currentspecification.types[field]](field,value,manipulator) - else + elseif trace_detail then report("%s %s %a in category %a for tag %a in dataset %a","unknown","field",field,category,tag,name) end - else + elseif trace_detail then report("%s %s %a in category %a for tag %a in dataset %a","invalid","field",field,category,tag,name) end else @@ -1105,10 +1082,10 @@ do local value = details[field] if value then typesetters[currentspecification.types[field]](field,value,manipulator) - else + elseif trace_detail then report("%s %s %a in category %a for tag %a in dataset %a","unknown","detail",field,category,tag,name) end - else + elseif trace_detail then report("%s %s %a in category %a for tag %a in dataset %a","invalid","detail",field,category,tag,name) end else @@ -1157,392 +1134,410 @@ function publications.singularorplural(singular,plural) end end -local patterns = { "publ-imp-%s.mkvi", "publ-imp-%s.mkiv", "publ-imp-%s.tex" } +-- loading -local function failure(name) - report("unknown library %a",name) -end +do -local function action(name,foundname) - context.input(foundname) -end + local patterns = { "publ-imp-%s.mkvi", "publ-imp-%s.mkiv", "publ-imp-%s.tex" } + + local function failure(name) + report("unknown library %a",name) + end + + local function action(name,foundname) + context.input(foundname) + end + + function commands.loadbtxdefinitionfile(name) -- a more specific name + commands.uselibrary { + name = string.gsub(name,"^publ%-",""), + patterns = patterns, + action = action, + failure = failure, + onlyonce = true, + } + end -function commands.loadbtxdefinitionfile(name) -- a more specific name - commands.uselibrary { - name = gsub(name,"^publ%-",""), - patterns = patterns, - action = action, - failure = failure, - onlyonce = true, - } end --- lists: +-- lists -publications.lists = publications.lists or { } -local lists = publications.lists +do -local context = context -local structures = structures + publications.lists = publications.lists or { } + local lists = publications.lists -local references = structures.references -local sections = structures.sections + local context = context + local structures = structures --- per rendering + local references = structures.references + local sections = structures.sections -local renderings = { } --- per dataset + -- per rendering -setmetatableindex(renderings,function(t,k) - local v = { - list = { }, - done = { }, - alldone = { }, - used = { }, - registered = { }, - ordered = { }, - shorts = { }, - method = v_none, - texts = setmetatableindex("table"), - currentindex = 0, - } - t[k] = v - return v -end) + local renderings = { } --- per dataset --- helper - --- local function sortedtags(dataset,list,sorttype) --- local luadata = datasets[dataset].luadata --- local valid = { } --- for i=1,#list do --- local tag = list[i] --- local entry = luadata[tag] --- if entry then --- local key = entry[sorttype] --- if key then --- valid[#valid+1] = { --- tag = tag, --- split = sortsplitter(sortstripper(key)) --- } --- end --- end --- end --- if #valid == 0 or #valid ~= #list then --- return list --- else --- sorters.sort(valid,basicsorter) --- for i=1,#valid do --- valid[i] = valid[i].tag --- end --- return valid --- end --- end --- --- if sorttype and sorttype ~= "" then --- tags = sortedtags(dataset,tags,sorttype) --- end + setmetatableindex(renderings,function(t,k) + local v = { + list = { }, + done = { }, + alldone = { }, + used = { }, + registered = { }, + ordered = { }, + shorts = { }, + method = v_none, + texts = setmetatableindex("table"), + currentindex = 0, + } + t[k] = v + return v + end) --- why shorts vs tags: only for sorting + -- helper + + -- local function sortedtags(dataset,list,sorttype) + -- local luadata = datasets[dataset].luadata + -- local valid = { } + -- for i=1,#list do + -- local tag = list[i] + -- local entry = luadata[tag] + -- if entry then + -- local key = entry[sorttype] + -- if key then + -- valid[#valid+1] = { + -- tag = tag, + -- split = sortsplitter(sortstripper(key)) + -- } + -- end + -- end + -- end + -- if #valid == 0 or #valid ~= #list then + -- return list + -- else + -- sorters.sort(valid,basicsorter) + -- for i=1,#valid do + -- valid[i] = valid[i].tag + -- end + -- return valid + -- end + -- end + -- + -- if sorttype and sorttype ~= "" then + -- tags = sortedtags(dataset,tags,sorttype) + -- end -function lists.register(dataset,tag,short) -- needs checking now that we split - local r = renderings[dataset] - if not short or short == "" then - short = tag - end - if trace then - report("registering publication entry %a with shortcut %a",tag,short) + -- why shorts vs tags: only for sorting + + function lists.register(dataset,tag,short) -- needs checking now that we split + local r = renderings[dataset] + if not short or short == "" then + short = tag + end + if trace then + report("registering publication entry %a with shortcut %a",tag,short) + end + local top = #r.registered + 1 + -- do we really need these + r.registered[top] = tag + r.ordered [tag] = top + r.shorts [tag] = short end - local top = #r.registered + 1 - -- do we really need these - r.registered[top] = tag - r.ordered [tag] = top - r.shorts [tag] = short -end -function lists.nofregistered(dataset) - return #renderings[dataset].registered -end + function lists.nofregistered(dataset) + return #renderings[dataset].registered + end -local function validkeyword(dataset,tag,keyword) - local ds = datasets[dataset] - if not ds then - report("unknown dataset %a",dataset) - return - end - local dt = ds.details[tag] - if not dt then - report("no details for tag %a",tag) - return - end - local kw = dt.keyword - if kw then - for k in next, keyword do - if kw[k] then - return true + local function validkeyword(dataset,entry,keyword) + local kw = fastget(dataset,entry,"keywords") -- hard coded for the moment + if kw then + for k in next, keyword do + if kw[k] then + return true + end end end end -end -local function registerpage(pages,tag,result,listindex) - local p = pages[tag] - local r = result[listindex].references - if p then - local last = p[#p][2] - local real = last.realpage - if real ~= r.realpage then - p[#p+1] = { listindex, r } + local function registerpage(pages,tag,result,listindex) + local p = pages[tag] + local r = result[listindex].references + if p then + local last = p[#p][2] + local real = last.realpage + if real ~= r.realpage then + p[#p+1] = { listindex, r } + end + else + pages[tag] = { { listindex, r } } end - else - pages[tag] = { { listindex, r } } end -end -local methods = { } -lists.methods = methods + local methods = { } + lists.methods = methods -methods[v_dataset] = function(dataset,rendering,keyword) - -- why only once unless criterium=all? - local luadata = datasets[dataset].luadata - local list = rendering.list - for tag, data in sortedhash(luadata) do - if not keyword or validkeyword(dataset,tag,keyword) then - list[#list+1] = { tag, false, 0, false, false } + methods[v_dataset] = function(dataset,rendering,keyword) + -- why only once unless criterium=all? + local current = datasets[dataset] + local luadata = current.luadata + local list = rendering.list + for tag, data in sortedhash(luadata) do + if not keyword or validkeyword(dataset,data,keyword) then + list[#list+1] = { tag, false, 0, false, false } + end end end -end -methods[v_force] = function (dataset,rendering,keyword) - -- only for checking, can have duplicates, todo: collapse page numbers, although - -- we then also needs deferred writes - local result = structures.lists.filter(rendering.specification) or { } - local list = rendering.list - for listindex=1,#result do - local r = result[listindex] - local u = r.userdata - if u and u.btxset == dataset then - local tag = u.btxref - if tag and (not keyword or validkeyword(dataset,tag,keyword)) then - list[#list+1] = { tag, listindex, 0, u, u.btxint } + methods[v_force] = function (dataset,rendering,keyword) + -- only for checking, can have duplicates, todo: collapse page numbers, although + -- we then also needs deferred writes + local result = structures.lists.filter(rendering.specification) or { } + local list = rendering.list + local current = datasets[dataset] + local luadata = current.luadata + for listindex=1,#result do + local r = result[listindex] + local u = r.userdata + if u and u.btxset == dataset then + local tag = u.btxref + if tag and (not keyword or validkeyword(dataset,luadata[tag],keyword)) then + list[#list+1] = { tag, listindex, 0, u, u.btxint } + end end end - end - lists.result = result -end - --- local : if tag and done[tag] ~= section then ... --- global : if tag and not alldone[tag] and done[tag] ~= section then ... - -methods[v_local] = function(dataset,rendering,keyword) - local result = structures.lists.filter(rendering.specification) or { } - local section = sections.currentid() - local list = rendering.list - local repeated = rendering.repeated == v_yes - local r_done = rendering.done - local r_alldone = rendering.alldone - local done = repeated and { } or r_done - local alldone = repeated and { } or r_alldone - local doglobal = rendering.method == v_global - local traced = { } -- todo: only if interactive (backlinks) or when tracing - local pages = { } - for listindex=1,#result do - local r = result[listindex] - local u = r.userdata - if u and u.btxset == dataset then - local tag = u.btxref - if not tag then - -- problem - elseif done[tag] == section then -- a bit messy for global and all and so - -- skip - elseif doglobal and alldone[tag] then - -- skip - elseif not keyword or validkeyword(dataset,tag,keyword) then - if traced then - local l = traced[tag] - if l then - l[#l+1] = u.btxint + lists.result = result + end + + -- local : if tag and done[tag] ~= section then ... + -- global : if tag and not alldone[tag] and done[tag] ~= section then ... + + methods[v_local] = function(dataset,rendering,keyword) + local result = structures.lists.filter(rendering.specification) or { } + local section = sections.currentid() + local list = rendering.list + local repeated = rendering.repeated == v_yes + local r_done = rendering.done + local r_alldone = rendering.alldone + local done = repeated and { } or r_done + local alldone = repeated and { } or r_alldone + local doglobal = rendering.method == v_global + local traced = { } -- todo: only if interactive (backlinks) or when tracing + local pages = { } + local current = datasets[dataset] + local luadata = current.luadata + for listindex=1,#result do + local r = result[listindex] + local u = r.userdata + if u and u.btxset == dataset then + local tag = u.btxref + if not tag then + -- problem + elseif done[tag] == section then -- a bit messy for global and all and so + -- skip + elseif doglobal and alldone[tag] then + -- skip + elseif not keyword or validkeyword(dataset,luadata[tag],keyword) then + if traced then + local l = traced[tag] + if l then + l[#l+1] = u.btxint + else + local l = { tag, listindex, 0, u, u.btxint } + list[#list+1] = l + traced[tag] = l + end else - local l = { tag, listindex, 0, u, u.btxint } - list[#list+1] = l - traced[tag] = l + done[tag] = section + alldone[tag] = true + list[#list+1] = { tag, listindex, 0, u, u.btxint } end - else - done[tag] = section - alldone[tag] = true - list[#list+1] = { tag, listindex, 0, u, u.btxint } end + registerpage(pages,tag,result,listindex) end - registerpage(pages,tag,result,listindex) end - end - if traced then - for tag in next, traced do - done[tag] = section - alldone[tag] = true + if traced then + for tag in next, traced do + done[tag] = section + alldone[tag] = true + end end + lists.result = result + structures.lists.result = result + rendering.pages = pages -- or list.pages + -- inspect(pages) + end + + methods[v_global] = methods[v_local] + + function lists.collectentries(specification) + local dataset = specification.btxdataset + if not dataset then + return + end + local rendering = renderings[dataset] + if not rendering then + return + end + local method = specification.method or v_none + local ignored = specification.ignored or "" + rendering.method = method + rendering.ignored = ignored ~= "" and settings_to_set(ignored) or nil + rendering.list = { } + rendering.done = { } + rendering.sorttype = specification.sorttype or v_default + rendering.criterium = specification.criterium or v_none + rendering.repeated = specification.repeated or v_no + rendering.specification = specification + local filtermethod = methods[method] + if not filtermethod then + return + end + lists.result = { } -- kind of reset + local keyword = specification.keyword + if keyword and keyword ~= "" then + keyword = settings_to_set(keyword) + else + keyword = nil + end + filtermethod(dataset,rendering,keyword) end - lists.result = result - structures.lists.result = result - rendering.pages = pages -- or list.pages - -- inspect(pages) -end - -methods[v_global] = methods[v_local] - -function lists.collectentries(specification) - local dataset = specification.btxdataset - if not dataset then - return - end - local rendering = renderings[dataset] - if not rendering then - return - end - local method = specification.method or v_none - local ignored = specification.ignored or "" - rendering.method = method - rendering.ignored = ignored ~= "" and settings_to_set(ignored) or nil - rendering.list = { } - rendering.done = { } - rendering.sorttype = specification.sorttype or v_default - rendering.criterium = specification.criterium or v_none - rendering.repeated = specification.repeated or v_no - rendering.specification = specification - local filtermethod = methods[method] - if not filtermethod then - return - end - lists.result = { } -- kind of reset - local keyword = specification.keyword - if keyword and keyword ~= "" then - keyword = settings_to_set(keyword) - else - keyword = nil - end - filtermethod(dataset,rendering,keyword) -end - --- experiment -local splitspec = lpeg.splitat(S(":.")) -local splitter = sorters.splitters.utf -local strip = sorters.strip + -- experiment -local function newsplitter(splitter) - return setmetatableindex({},function(t,k) -- could be done in the sorter but seldom that many shared - local v = splitter(k,true) -- in other cases - t[k] = v - return v - end) -end + local splitspec = lpeg.splitat(S(":.")) + local splitter = sorters.splitters.utf + local strip = sorters.strip -local template = [[ - local strip = sorters.strip - local writers = publications.writers - return function(entry,detail,splitted,i) -- snippets - return { - index = i, - split = { %s, splitted[tostring(i)] } - } + local function newsplitter(splitter) + return setmetatableindex({},function(t,k) -- could be done in the sorter but seldom that many shared + local v = splitter(k,true) -- in other cases + t[k] = v + return v + end) end -]] - -local function byspec(dataset,list,method) -- todo: yearsuffix - local luadata = datasets[dataset].luadata - local details = datasets[dataset].details - local result = { } - local splitted = newsplitter(splitter) -- saves mem - -- local snippets = { } -- saves mem - local fields = settings_to_array(method) - for i=1,#fields do - local f = settings_to_array(fields[i]) - local r = { } - for i=1,#f do - local a, b = lpegmatch(splitspec,f[i]) - if b then - if a == "detail" or a == "entry" then - local t = currentspecification.types[b] - local w = t and writers[t] - if w then - r[#r+1] = formatters["(%s.%s and writers[%q](%s.%s))"](a,b,t,a,b) - else - r[#r+1] = formatters["%s.%s"](a,b,a,b) + + local template = [[ + local strip = sorters.strip + local writers = publications.writers + return function(entry,detail,splitted,i) -- snippets + return { + index = i, + split = { %s, splitted[tostring(i)] } + } + end + ]] + + local function byspec(dataset,list,method) -- todo: yearsuffix + local luadata = datasets[dataset].luadata + local details = datasets[dataset].details + local result = { } + local splitted = newsplitter(splitter) -- saves mem + -- local snippets = { } -- saves mem + local fields = settings_to_array(method) + for i=1,#fields do + local f = settings_to_array(fields[i]) + local r = { } + for i=1,#f do + local a, b = lpegmatch(splitspec,f[i]) + if b then + if a == "detail" or a == "entry" then + local t = currentspecification.types[b] + local w = t and writers[t] + if w then + r[#r+1] = formatters["(%s.%s and writers[%q](%s.%s))"](a,b,t,a,b) + else + r[#r+1] = formatters["%s.%s"](a,b,a,b) + end end + elseif a then + r[#r+1] = formatters["%s"](a) end - elseif a then - r[#r+1] = formatters["%s"](a) end - end - r[#r+1] = '""' - fields[i] = "splitted[strip(" .. concat(r," or ") .. ")]" - end - local action = formatters[template](concat(fields,", ")) - local prepare = loadstring(action) - if prepare then - prepare = prepare() - local dummy = { } - for i=1,#list do - -- either { tag, tag, ... } or { { tag, index }, { tag, index } } - local li = list[i] - local tag = type(li) == "string" and li or li[1] - local entry = luadata[tag] - local detail = details[tag] - if entry and detail then - result[i] = prepare(entry,detail,splitted,i) -- ,snippets) - else - result[i] = prepare(dummy,dummy,splitted,i) -- ,snippets) + r[#r+1] = '""' + fields[i] = "splitted[strip(" .. concat(r," or ") .. ")]" + end + local action = formatters[template](concat(fields,", ")) + local prepare = loadstring(action) + if prepare then + prepare = prepare() + local dummy = { } + for i=1,#list do + -- either { tag, tag, ... } or { { tag, index }, { tag, index } } + local li = list[i] + local tag = type(li) == "string" and li or li[1] + local entry = luadata[tag] + local detail = details[tag] + if entry and detail then + result[i] = prepare(entry,detail,splitted,i) -- ,snippets) + else + result[i] = prepare(dummy,dummy,splitted,i) -- ,snippets) + end end end + return result end - return result -end -lists.sorters = { - [v_short] = function(dataset,rendering,list) - local shorts = rendering.shorts - local function compare(a,b) - local aa, bb = a and a[1], b and b[1] - if aa and bb then - aa, bb = shorts[aa], shorts[bb] - return aa and bb and aa < bb - end - return false - end - sort(list,compare) - end, - [v_reference] = function(dataset,rendering,list) - local function compare(a,b) - local aa, bb = a and a[1], b and b[1] - if aa and bb then - return aa and bb and aa < bb + lists.sorters = { + [v_short] = function(dataset,rendering,list) + local shorts = rendering.shorts + local function compare(a,b) + local aa, bb = a and a[1], b and b[1] + if aa and bb then + aa, bb = shorts[aa], shorts[bb] + return aa and bb and aa < bb + end + return false end - return false - end - sort(list,compare) - end, - [v_dataset] = function(dataset,rendering,list) - local function compare(a,b) - local aa, bb = a and a[1], b and b[1] - if aa and bb then - aa, bb = list[aa].index or 0, list[bb].index or 0 - return aa and bb and aa < bb + sort(list,compare) + end, + [v_reference] = function(dataset,rendering,list) + local function compare(a,b) + local aa, bb = a and a[1], b and b[1] + if aa and bb then + return aa and bb and aa < bb + end + return false end - return false - end - sort(list,compare) - end, - [v_default] = function(dataset,rendering,list,sorttype) -- experimental - if sorttype == "" or sorttype == v_default then + sort(list,compare) + end, + [v_dataset] = function(dataset,rendering,list) local function compare(a,b) - local aa, bb = a and a[3], b and b[3] + local aa, bb = a and a[1], b and b[1] if aa and bb then + aa, bb = list[aa].index or 0, list[bb].index or 0 return aa and bb and aa < bb end return false end sort(list,compare) - else - local valid = byspec(dataset,list,sorttype) + end, + [v_default] = function(dataset,rendering,list,sorttype) -- experimental + if sorttype == "" or sorttype == v_default then + local function compare(a,b) + local aa, bb = a and a[3], b and b[3] + if aa and bb then + return aa and bb and aa < bb + end + return false + end + sort(list,compare) + else + local valid = byspec(dataset,list,sorttype) + if #valid == 0 or #valid ~= #list then + -- nothing to sort + else + -- if needed we can wrap compare and use the list directly but this is cleaner + sorters.sort(valid,sortcomparer) + for i=1,#valid do + local v = valid[i] + valid[i] = list[v.index] + end + return valid + end + end + end, + [v_author] = function(dataset,rendering,list) + local valid = publications.authors.sorters.author(dataset,list) if #valid == 0 or #valid ~= #list then -- nothing to sort else @@ -1554,1181 +1549,1176 @@ lists.sorters = { end return valid end - end - end, - [v_author] = function(dataset,rendering,list) - local valid = publications.authors.sorters.author(dataset,list) - if #valid == 0 or #valid ~= #list then - -- nothing to sort - else - -- if needed we can wrap compare and use the list directly but this is cleaner - sorters.sort(valid,sortcomparer) - for i=1,#valid do - local v = valid[i] - valid[i] = list[v.index] - end - return valid - end - end, -} + end, + } --- for determining width - -local lastreferencenumber = 0 -- document wide - -function lists.prepareentries(dataset) - local rendering = renderings[dataset] - local list = rendering.list - local used = rendering.used - local forceall = rendering.criterium == v_all - local repeated = rendering.repeated == v_yes - local sorttype = rendering.sorttype or v_default - local sorter = lists.sorters[sorttype] or lists.sorters[v_default] - local current = datasets[dataset] - local luadata = current.luadata - local details = current.details - local newlist = { } - for i=1,#list do - local li = list[i] - local tag = li[1] - local entry = luadata[tag] - if entry then - if forceall or repeated or not used[tag] then - newlist[#newlist+1] = li - -- already here: - if not repeated then - used[tag] = true -- beware we keep the old state (one can always use criterium=all) - end - local detail = details[tag] - if detail then - local referencenumber = detail.referencenumber - if not referencenumber then - lastreferencenumber = lastreferencenumber + 1 - referencenumber = lastreferencenumber - detail.referencenumber = lastreferencenumber + -- for determining width + + local lastreferencenumber = 0 -- document wide + + function lists.prepareentries(dataset) + local rendering = renderings[dataset] + local list = rendering.list + local used = rendering.used + local forceall = rendering.criterium == v_all + local repeated = rendering.repeated == v_yes + local sorttype = rendering.sorttype or v_default + local sorter = lists.sorters[sorttype] or lists.sorters[v_default] + local current = datasets[dataset] + local luadata = current.luadata + local details = current.details + local newlist = { } + for i=1,#list do + local li = list[i] + local tag = li[1] + local entry = luadata[tag] + if entry then + if forceall or repeated or not used[tag] then + newlist[#newlist+1] = li + -- already here: + if not repeated then + used[tag] = true -- beware we keep the old state (one can always use criterium=all) + end + local detail = details[tag] + if detail then + local referencenumber = detail.referencenumber + if not referencenumber then + lastreferencenumber = lastreferencenumber + 1 + referencenumber = lastreferencenumber + detail.referencenumber = lastreferencenumber + end + li[3] = referencenumber + else + report("missing details for tag %a in dataset %a (enhanced: %s)",tag,dataset,current.enhanced and "yes" or "no") + -- weird, this shouldn't happen .. all have a detail + lastreferencenumber = lastreferencenumber + 1 + details[tag] = { referencenumber = lastreferencenumber } + li[3] = lastreferencenumber end - li[3] = referencenumber - else - report("missing details for tag %a in dataset %a (enhanced: %s)",tag,dataset,current.enhanced and "yes" or "no") - -- weird, this shouldn't happen .. all have a detail - lastreferencenumber = lastreferencenumber + 1 - details[tag] = { referencenumber = lastreferencenumber } - li[3] = lastreferencenumber end end end + rendering.list = type(sorter) == "function" and sorter(dataset,rendering,newlist,sorttype) or newlist end - rendering.list = type(sorter) == "function" and sorter(dataset,rendering,newlist,sorttype) or newlist -end -function lists.fetchentries(dataset) - local rendering = renderings[dataset] - local list = rendering.list - for i=1,#list do - local li = list[i] - ctx_btxsettag(li[1]) - ctx_btxsetnumber(li[3]) - ctx_btxchecklistentry() + function lists.fetchentries(dataset) + local rendering = renderings[dataset] + local list = rendering.list + for i=1,#list do + local li = list[i] + ctx_btxsettag(li[1]) + ctx_btxsetnumber(li[3]) + ctx_btxchecklistentry() + end end -end --- for rendering - --- setspecification - -function commands.btxflushpages(dataset,tag) - -- todo: interaction - local rendering = renderings[dataset] - local pages = rendering.pages[tag] - if not pages then - return - end - local nofpages = #pages - if nofpages == 0 then - return - end - local first_p = nil - local first_r = nil - local last_p = nil - local last_r = nil - local ranges = { } - local nofdone = 0 - local function flush() - if last_r and first_r ~= last_r then - ranges[#ranges+1] = { first_p, last_p } - else - ranges[#ranges+1] = { first_p } - end - end - for i=1,nofpages do - local next_p = pages[i] - local next_r = next_p[2].realpage - if not first_r then - first_p = next_p - first_r = next_r - elseif last_r + 1 == next_r then - -- continue - elseif first_r then - flush() - first_p = next_p - first_r = next_r + -- for rendering + + -- setspecification + + function commands.btxflushpages(dataset,tag) + -- todo: interaction + local rendering = renderings[dataset] + local pages = rendering.pages[tag] + if not pages then + return end - last_p = next_p - last_r = next_r - end - if first_r then - flush() - end - local nofranges = #ranges - for i=1,nofranges do - local r = ranges[i] - ctx_btxsetconcat(concatstate(i,nofranges)) - local first, last = r[1], r[2] - ctx_btxsetfirstinternal(first[2].internal) - ctx_btxsetfirstpage(first[1]) - if last then - ctx_btxsetlastinternal(last[2].internal) - ctx_btxsetlastpage(last[1]) + local nofpages = #pages + if nofpages == 0 then + return end - ctx_btxpagesetup() - end -end - -function lists.flushentries(dataset,textmode) - local rendering = renderings[dataset] - local list = rendering.list - local luadata = datasets[dataset].luadata - -- maybe a startflushing here - ignoredfields = rendering.ignored or { } - -- - for i=1,#list do - local li = list[i] - local tag = li[1] - local n = li[3] - local entry = luadata[tag] - local combined = entry.combined - local language = entry.language - if combined then - ctx_btxsetcombis(concat(combined,",")) - end - ctx_btxsetcategory(entry.category or "unknown") - ctx_btxsettag(tag) - ctx_btxsetnumber(n) - if language then - ctx_btxsetlanguage(language) - end - local bl = li[5] - if bl and bl ~= "" then - ctx_btxsetbacklink(bl) - ctx_btxsetbacktrace(concat(li," ",5)) - local uc = citetolist[tonumber(bl)] - if uc then - ctx_btxsetinternal(uc.references.internal or "") + local first_p = nil + local first_r = nil + local last_p = nil + local last_r = nil + local ranges = { } + local nofdone = 0 + local function flush() + if last_r and first_r ~= last_r then + ranges[#ranges+1] = { first_p, last_p } + else + ranges[#ranges+1] = { first_p } end - else - -- nothing end - local userdata = li[4] - if userdata then - local b = userdata.btxbtx - local a = userdata.btxatx - if b then - ctx_btxsetbefore(b) - end - if a then - ctx_btxsetafter(a) + for i=1,nofpages do + local next_p = pages[i] + local next_r = next_p[2].realpage + if not first_r then + first_p = next_p + first_r = next_r + elseif last_r + 1 == next_r then + -- continue + elseif first_r then + flush() + first_p = next_p + first_r = next_r end + last_p = next_p + last_r = next_r end - rendering.userdata = userdata - if textmode then - ctx_btxhandlelisttextentry() - else - ctx_btxhandlelistentry() + if first_r then + flush() + end + local nofranges = #ranges + for i=1,nofranges do + local r = ranges[i] + ctx_btxsetconcat(concatstate(i,nofranges)) + local first, last = r[1], r[2] + ctx_btxsetfirstinternal(first[2].internal) + ctx_btxsetfirstpage(first[1]) + if last then + ctx_btxsetlastinternal(last[2].internal) + ctx_btxsetlastpage(last[1]) + end + ctx_btxpagesetup() end end - context(function() - -- wrapup - ignoredfields = nil - setspecification(false) - end) -end -local function getuserdata(dataset,key) - local rendering = renderings[dataset] - if rendering then - local userdata = rendering.userdata - if userdata then - local value = userdata[key] - if value and value ~= "" then - return value + function lists.flushentries(dataset,textmode) + local rendering = renderings[dataset] + local list = rendering.list + local luadata = datasets[dataset].luadata + -- maybe a startflushing here + ignoredfields = rendering.ignored or { } + -- + for i=1,#list do + local li = list[i] + local tag = li[1] + local n = li[3] + local entry = luadata[tag] + local combined = entry.combined + local language = entry.language + if combined then + ctx_btxsetcombis(concat(combined,",")) + end + ctx_btxsetcategory(entry.category or "unknown") + ctx_btxsettag(tag) + ctx_btxsetnumber(n) + if language then + ctx_btxsetlanguage(language) + end + local bl = li[5] + if bl and bl ~= "" then + ctx_btxsetbacklink(bl) + ctx_btxsetbacktrace(concat(li," ",5)) + local uc = citetolist[tonumber(bl)] + if uc then + ctx_btxsetinternal(uc.references.internal or "") + end + else + -- nothing + end + local userdata = li[4] + if userdata then + local b = userdata.btxbtx + local a = userdata.btxatx + if b then + ctx_btxsetbefore(b) + end + if a then + ctx_btxsetafter(a) + end + end + rendering.userdata = userdata + if textmode then + ctx_btxhandlelisttextentry() + else + ctx_btxhandlelistentry() + end + end + context(function() + -- wrapup + ignoredfields = nil + setspecification(false) + end) + end + + local function getuserdata(dataset,key) + local rendering = renderings[dataset] + if rendering then + local userdata = rendering.userdata + if userdata then + local value = userdata[key] + if value and value ~= "" then + return value + end end end end -end -lists.uservariable = getuserdata + lists.uservariable = getuserdata -function commands.btxuservariable(dataset,key) - local value = getuserdata(dataset,key) - if value then - context(value) + function commands.btxuservariable(dataset,key) + local value = getuserdata(dataset,key) + if value then + context(value) + end end -end -function commands.btxdoifelseuservariable(dataset,key) - if getuserdata(dataset,key) then - ctx_firstoftwoarguments() - else - ctx_secondoftwoarguments() + function commands.btxdoifelseuservariable(dataset,key) + if getuserdata(dataset,key) then + ctx_firstoftwoarguments() + else + ctx_secondoftwoarguments() + end end -end -function lists.filterall(dataset) - local r = renderings[dataset] - local list = r.list - local registered = r.registered - for i=1,#registered do - list[i] = { registered[i], i, 0, false, false } + function lists.filterall(dataset) + local r = renderings[dataset] + local list = r.list + local registered = r.registered + for i=1,#registered do + list[i] = { registered[i], i, 0, false, false } + end end -end - -commands.btxresolvelistreference = lists.resolve -commands.btxaddtolist = lists.addentry -commands.btxcollectlistentries = lists.collectentries -commands.btxpreparelistentries = lists.prepareentries -commands.btxfetchlistentries = lists.fetchentries -commands.btxflushlistentries = lists.flushentries -commands.btxflushlistentry = lists.flushentry -local citevariants = { } -publications.citevariants = citevariants + commands.btxresolvelistreference = lists.resolve + commands.btxaddtolist = lists.addentry + commands.btxcollectlistentries = lists.collectentries + commands.btxpreparelistentries = lists.prepareentries + commands.btxfetchlistentries = lists.fetchentries + commands.btxflushlistentries = lists.flushentries + commands.btxflushlistentry = lists.flushentry -function commands.btxhandlecite(specification) - local dataset = specification.dataset or "" -- standard - local reference = specification.reference - local variant = specification.variant or defaultvariant - if not reference or reference == "" then - return - end - -- - specification.variant = variant - specification.compress = specification.compress == v_yes - specification.markentry = specification.markentry ~= false - -- - local prefix, rest = lpegmatch(prefixsplitter,reference) - if prefix and rest then - specification.dataset = prefix - specification.reference = rest - end - -- - -- - if trace_cite then - report_cite("inject, dataset: %s, tag: %s, variant: %s, compressed", - specification.dataset or "-", - specification.reference, - specification.variant - ) - end - -- - ctx_setvalue("currentbtxdataset",dataset) - -- - citevariants[variant](specification) -- we always fall back on default end +do -function commands.btxhandlenocite(specification) - local dataset = specification.dataset or "" -- standard - local reference = specification.reference - if not reference or reference == "" then - return - end - -- - local markentry = specification.markentry ~= false - local internal = specification.internal or "" - -- - local prefix, rest = lpegmatch(prefixsplitter,reference) - if rest then - dataset = prefix - reference = rest - end - -- - if trace_cite then - report_cite("mark, dataset: %s, tags: %s",dataset or "-",reference) - end - -- - local reference = publications.parenttag(dataset,reference) - -- - local found, todo, list = findallused(dataset,reference,internal) - -- - tobemarked = markentry and todo - if found and tobemarked then - flushmarked(dataset,list) - commands.flushmarked() -- here (could also be done in caller) - end -end - --- function commands.btxcitevariant(dataset,block,tags,variant) -- uses? specification ? --- local action = citevariants[variant] --- if action then --- action(dataset,tags,variant) --- end --- end - --- sorter - -local keysorter = function(a,b) return a.sortkey < b.sortkey end - --- local suffix = 0 --- local function setsuffix(entry,suffix,sortfld) --- entry.suffix = suffix --- local dataset = datasets[entry.dataset] --- if dataset then --- local suffixes = dataset.suffixes[entry.tag] --- if suffixes then --- suffixes[sortfld] = suffix --- else --- dataset.suffixes[entry.tag] = { [sortfld] = suffix } --- end --- end --- end --- for i=1,#source do --- local entry = source[i] --- local sortfld = entry.sortfld --- if sortfld then --- local value = entry.sortkey --- if value == oldvalue then --- if suffix == 0 then --- suffix = 1 --- local entry = source[i-1] --- setsuffix(entry,suffix,sortfld) --- end --- suffix = suffix + 1 --- setsuffix(entry,suffix,sortfld) --- else --- oldvalue = value --- suffix = 0 --- end --- else --- break --- end --- end - -local function compresslist(source) - for i=1,#source do - if type(source[i].sortkey) ~= "number" then - return source - end - end - local first, last, firstr, lastr - local target, noftarget, tags = { }, 0, { } - sort(source,keysorter) - local oldvalue = nil - local function flushrange() - noftarget = noftarget + 1 - if last > first + 1 then - target[noftarget] = { - first = firstr, - last = lastr, - tags = tags, - } - else - target[noftarget] = firstr - if last > first then - noftarget = noftarget + 1 - target[noftarget] = lastr + local citevariants = { } + publications.citevariants = citevariants + + function commands.btxhandlecite(specification) + local dataset = specification.dataset or "" -- standard + local reference = specification.reference + local variant = specification.variant or defaultvariant + if not reference or reference == "" then + return + end + -- + specification.variant = variant + specification.compress = specification.compress == v_yes + specification.markentry = specification.markentry ~= false + -- + local prefix, rest = lpegmatch(prefixsplitter,reference) + if prefix and rest then + specification.dataset = prefix + specification.reference = rest + end + -- + -- + if trace_cite then + report_cite("inject, dataset: %s, tag: %s, variant: %s, compressed", + specification.dataset or "-", + specification.reference, + specification.variant + ) + end + -- + ctx_setvalue("currentbtxdataset",dataset) + -- + citevariants[variant](specification) -- we always fall back on default + end + + + function commands.btxhandlenocite(specification) + local dataset = specification.dataset or "" -- standard + local reference = specification.reference + if not reference or reference == "" then + return + end + -- + local markentry = specification.markentry ~= false + local internal = specification.internal or "" + -- + local prefix, rest = lpegmatch(prefixsplitter,reference) + if rest then + dataset = prefix + reference = rest + end + -- + if trace_cite then + report_cite("mark, dataset: %s, tags: %s",dataset or "-",reference) + end + -- + local reference = publications.parenttag(dataset,reference) + -- + local found, todo, list = findallused(dataset,reference,internal) + -- + tobemarked = markentry and todo + if found and tobemarked then + flushmarked(dataset,list) + commands.flushmarked() -- here (could also be done in caller) + end + end + + -- function commands.btxcitevariant(dataset,block,tags,variant) -- uses? specification ? + -- local action = citevariants[variant] + -- if action then + -- action(dataset,tags,variant) + -- end + -- end + + -- sorter + + local keysorter = function(a,b) return a.sortkey < b.sortkey end + + -- local suffix = 0 + -- local function setsuffix(entry,suffix,sortfld) + -- entry.suffix = suffix + -- local dataset = datasets[entry.dataset] + -- if dataset then + -- local suffixes = dataset.suffixes[entry.tag] + -- if suffixes then + -- suffixes[sortfld] = suffix + -- else + -- dataset.suffixes[entry.tag] = { [sortfld] = suffix } + -- end + -- end + -- end + -- for i=1,#source do + -- local entry = source[i] + -- local sortfld = entry.sortfld + -- if sortfld then + -- local value = entry.sortkey + -- if value == oldvalue then + -- if suffix == 0 then + -- suffix = 1 + -- local entry = source[i-1] + -- setsuffix(entry,suffix,sortfld) + -- end + -- suffix = suffix + 1 + -- setsuffix(entry,suffix,sortfld) + -- else + -- oldvalue = value + -- suffix = 0 + -- end + -- else + -- break + -- end + -- end + + local function compresslist(source) + for i=1,#source do + if type(source[i].sortkey) ~= "number" then + return source end end - tags = { } - end - for i=1,#source do - local entry = source[i] - local current = entry.sortkey - if not first then - first, last, firstr, lastr = current, current, entry, entry - elseif current == last + 1 then - last, lastr = current, entry - else + local first, last, firstr, lastr + local target, noftarget, tags = { }, 0, { } + sort(source,keysorter) + local oldvalue = nil + local function flushrange() + noftarget = noftarget + 1 + if last > first + 1 then + target[noftarget] = { + first = firstr, + last = lastr, + tags = tags, + } + else + target[noftarget] = firstr + if last > first then + noftarget = noftarget + 1 + target[noftarget] = lastr + end + end + tags = { } + end + for i=1,#source do + local entry = source[i] + local current = entry.sortkey + if not first then + first, last, firstr, lastr = current, current, entry, entry + elseif current == last + 1 then + last, lastr = current, entry + else + flushrange() + first, last, firstr, lastr = current, current, entry, entry + end + tags[#tags+1] = entry.tag + end + if first and last then flushrange() - first, last, firstr, lastr = current, current, entry, entry end - tags[#tags+1] = entry.tag + return target end - if first and last then - flushrange() - end - return target -end --- local source = { --- { tag = "one", internal = 1, value = "foo", page = 1 }, --- { tag = "two", internal = 2, value = "bar", page = 2 }, --- { tag = "three", internal = 3, value = "gnu", page = 3 }, --- } --- --- local target = compresslist(source) - -local numberonly = R("09")^1 / tonumber + P(1)^0 -local f_missing = formatters["<%s>"] + -- local source = { + -- { tag = "one", internal = 1, value = "foo", page = 1 }, + -- { tag = "two", internal = 2, value = "bar", page = 2 }, + -- { tag = "three", internal = 3, value = "gnu", page = 3 }, + -- } + -- + -- local target = compresslist(source) --- maybe also sparse (e.g. pages) + local numberonly = R("09")^1 / tonumber + P(1)^0 + local f_missing = formatters["<%s>"] --- a bit redundant access to datasets + -- maybe also sparse (e.g. pages) -local function processcite(presets,specification) - -- - if specification then - setmetatableindex(specification,presets) - else - specification = presets - end - -- - local dataset = specification.dataset - local reference = specification.reference - local internal = specification.internal - local setup = specification.variant - local compress = specification.compress - local getter = specification.getter - local setter = specification.setter - local compressor = specification.compressor - -- + -- a bit redundant access to datasets - local reference = publications.parenttag(dataset,reference) - -- - local found, todo, list = findallused(dataset,reference,internal) - tobemarked = specification.markentry and todo - -- - if found and setup then - local source = { } - local badkey = false - local luadata = datasets[dataset].luadata - for i=1,#found do - local entry = found[i] - local tag = entry.userdata.btxref - -- we can probably move the test into the flush - -- local category = luadata[tag].category - -- if currentspecificationfields[category][setup] then - local internal = entry.references.internal - local data = setter(dataset,tag,entry,internal) - if compress and not compressor then - local sortkey = data.sortkey - if sortkey then - local key = lpegmatch(numberonly,sortkey) - if key then - data.sortkey = key + local function processcite(presets,specification) + -- + if specification then + setmetatableindex(specification,presets) + else + specification = presets + end + -- + local dataset = specification.dataset + local reference = specification.reference + local internal = specification.internal + local setup = specification.variant + local compress = specification.compress + local getter = specification.getter + local setter = specification.setter + local compressor = specification.compressor + -- + + local reference = publications.parenttag(dataset,reference) + -- + local found, todo, list = findallused(dataset,reference,internal) + tobemarked = specification.markentry and todo + -- + if found and setup then + local source = { } + local badkey = false + local luadata = datasets[dataset].luadata + for i=1,#found do + local entry = found[i] + local tag = entry.userdata.btxref + -- we can probably move the test into the flush + -- local category = luadata[tag].category + -- if currentspecificationfields[category][setup] then + local internal = entry.references.internal + local data = setter(dataset,tag,entry,internal) + if compress and not compressor then + local sortkey = data.sortkey + if sortkey then + local key = lpegmatch(numberonly,sortkey) + if key then + data.sortkey = key + else + badkey = true + end else badkey = true end + end + if type(data) == "table" then + source[#source+1] = data else - badkey = true + report("error in cite rendering %a",setup or "?") end - end - if type(data) == "table" then - source[#source+1] = data - else - report("error in cite rendering %a",setup or "?") - end - -- else - -- report("cite rendering %a is not available for %a",setup,category) - -- end - end + -- else + -- report("cite rendering %a is not available for %a",setup,category) + -- end + end - local lefttext = specification.lefttext - local righttext = specification.righttext - local before = specification.before - local after = specification.after + local lefttext = specification.lefttext + local righttext = specification.righttext + local before = specification.before + local after = specification.after - if lefttext and lefttext ~= "" then lefttext = settings_to_array(lefttext) end - if righttext and righttext ~= "" then righttext = settings_to_array(righttext) end - if before and before ~= "" then before = settings_to_array(before) end - if after and after ~= "" then after = settings_to_array(after) end + if lefttext and lefttext ~= "" then lefttext = settings_to_array(lefttext) end + if righttext and righttext ~= "" then righttext = settings_to_array(righttext) end + if before and before ~= "" then before = settings_to_array(before) end + if after and after ~= "" then after = settings_to_array(after) end - local function flush(i,n,entry,last) - local tag = entry.tag - local currentcitation = markcite(dataset,tag) - -- - ctx_btxstartcite() - ctx_btxsettag(tag) - -- - if lefttext then ctx_btxsetlefttext (lefttext [i] or #lefttext == 1 and lefttext [1] or "") end - if righttext then ctx_btxsetrighttext(righttext[i] or #righttext == 1 and righttext[1] or "") end - if before then ctx_btxsetbefore (before [i] or #before == 1 and before [1] or "") end - if after then ctx_btxsetafter (after [i] or #after == 1 and after [1] or "") end - -- - ctx_btxsetbacklink(currentcitation) - local bl = listtocite[currentcitation] - if bl then - -- we refer to a coming list entry - ctx_btxsetinternal(bl.references.internal or "") - else - -- we refer to a previous list entry - ctx_btxsetinternal(entry.internal or "") - end - local language = entry.language - if language then - ctx_btxsetlanguage(language) - end - if not getter(entry,last) then - ctx_btxsetfirst(f_missing(tag)) + local function flush(i,n,entry,last) + local tag = entry.tag + local currentcitation = markcite(dataset,tag) + -- + ctx_btxstartcite() + ctx_btxsettag(tag) + -- + if lefttext then ctx_btxsetlefttext (lefttext [i] or #lefttext == 1 and lefttext [1] or "") end + if righttext then ctx_btxsetrighttext(righttext[i] or #righttext == 1 and righttext[1] or "") end + if before then ctx_btxsetbefore (before [i] or #before == 1 and before [1] or "") end + if after then ctx_btxsetafter (after [i] or #after == 1 and after [1] or "") end + -- + ctx_btxsetbacklink(currentcitation) + local bl = listtocite[currentcitation] + if bl then + -- we refer to a coming list entry + ctx_btxsetinternal(bl.references.internal or "") + else + -- we refer to a previous list entry + ctx_btxsetinternal(entry.internal or "") + end + local language = entry.language + if language then + ctx_btxsetlanguage(language) + end + if not getter(entry,last) then + ctx_btxsetfirst(f_missing(tag)) + end + ctx_btxsetconcat(concatstate(i,n)) + ctx_btxcitesetup(setup) + ctx_btxstopcite() end - ctx_btxsetconcat(concatstate(i,n)) - ctx_btxcitesetup(setup) - ctx_btxstopcite() - end - if compress and not badkey then - local target = (compressor or compresslist)(source) - local nofcollected = #target - if nofcollected == 0 then - unknowncite(reference) - else - for i=1,nofcollected do - local entry = target[i] - local first = entry.first - if first then - flush(i,nofcollected,first,entry.last) - else - flush(i,nofcollected,entry) + if compress and not badkey then + local target = (compressor or compresslist)(source) + local nofcollected = #target + if nofcollected == 0 then + unknowncite(reference) + else + for i=1,nofcollected do + local entry = target[i] + local first = entry.first + if first then + flush(i,nofcollected,first,entry.last) + else + flush(i,nofcollected,entry) + end end end - end - else - local nofcollected = #source - if nofcollected == 0 then - unknowncite(reference) else - for i=1,nofcollected do - flush(i,nofcollected,source[i]) + local nofcollected = #source + if nofcollected == 0 then + unknowncite(reference) + else + for i=1,nofcollected do + flush(i,nofcollected,source[i]) + end end end end + if tobemarked then + flushmarked(dataset,list) + commands.flushmarked() -- here (could also be done in caller) + end end - if tobemarked then - flushmarked(dataset,list) - commands.flushmarked() -- here (could also be done in caller) - end -end -local function simplegetter(first,last,field) - local value = first[field] - if value then - ctx_btxsetfirst(value) - if last then - ctx_btxsetsecond(last[field]) + local function simplegetter(first,last,field) + local value = first[field] + if value then + ctx_btxsetfirst(value) + if last then + ctx_btxsetsecond(last[field]) + end + return true end - return true end -end -local setters = setmetatableindex({},function(t,k) - local v = function(dataset,tag,entry,internal) - local value = getfield(dataset,tag,k) - return { - tag = tag, - internal = internal, - [k] = value, - sortkey = value, - sortfld = k, - } - end - t[k] = v - return v -end) + local setters = setmetatableindex({},function(t,k) + local v = function(dataset,tag,entry,internal) + local value = getfield(dataset,tag,k) + return { + tag = tag, + internal = internal, + [k] = value, + sortkey = value, + sortfld = k, + } + end + t[k] = v + return v + end) -local getters = setmetatableindex({},function(t,k) - local v = function(first,last) - return simplegetter(first,last,k) - end - t[k] = v - return v -end) + local getters = setmetatableindex({},function(t,k) + local v = function(first,last) + return simplegetter(first,last,k) + end + t[k] = v + return v + end) --- todo: just a sort key and then fetch normal by fieldname + -- todo: just a sort key and then fetch normal by fieldname --- setmetatableindex(citevariants,function(t,k) --- local p = registeredcitevariants[k] --- local v = p and p ~= k and rawget(t,p) or defaultvariant --- t[k] = v --- return v --- end) + -- setmetatableindex(citevariants,function(t,k) + -- local p = registeredcitevariants[k] + -- local v = p and p ~= k and rawget(t,p) or defaultvariant + -- t[k] = v + -- return v + -- end) -setmetatableindex(citevariants,function(t,k) - local p = registeredcitevariants[k] - local v = nil - if p and p ~= "" then - v = rawget(t,p) - end - if not v then - p = defaultvariant or "default" - v = rawget(t,p) + setmetatableindex(citevariants,function(t,k) + local p = registeredcitevariants[k] + local v = nil + if p and p ~= "" then + v = rawget(t,p) + end + if not v then + p = defaultvariant or "default" + v = rawget(t,p) + end + report_cite("variant %a falls back on %a",k,p) + t[k] = v + return v + end) + + function citevariants.default(presets) -- no longer used + local variant = presets.variant + processcite(presets,{ + setter = setters[variant], + getter = getters[variant], + }) end - report_cite("variant %a falls back on %a",k,p) - t[k] = v - return v -end) -function citevariants.default(presets) -- no longer used - local variant = presets.variant - processcite(presets,{ - setter = setters[variant], - getter = getters[variant], - }) -end + -- entry --- entry + do -do + local function setter(dataset,tag,entry,internal) + return { + tag = tag, + internal = internal, + } + end - local function setter(dataset,tag,entry,internal) - return { - tag = tag, - internal = internal, - } - end + local function getter(first,last) -- last not used + ctx_btxsetfirst(first.tag) + end - local function getter(first,last) -- last not used - ctx_btxsetfirst(first.tag) - end + function citevariants.entry(presets) + processcite(presets,{ + compress = false, + -- variant = presets.variant or "entry", + setter = setter, + getter = getter, + }) + end - function citevariants.entry(presets) - processcite(presets,{ - compress = false, - -- variant = presets.variant or "entry", - setter = setter, - getter = getter, - }) end -end - --- short + -- short -do + do - local function setter(dataset,tag,entry,internal) - return { - tag = tag, - internal = internal, - short = getdetail(dataset,tag,"short"), - suffix = getdetail(dataset,tag,"suffix"), - } - end + local function setter(dataset,tag,entry,internal) + return { + tag = tag, + internal = internal, + short = getdetail(dataset,tag,"short"), + suffix = getdetail(dataset,tag,"suffix"), + } + end - local function getter(first,last) -- last not used - local short = first.short - if short then - local suffix = first.suffix - if suffix then - ctx_btxsetfirst(short .. suffix) - else - ctx_btxsetfirst(short) + local function getter(first,last) -- last not used + local short = first.short + if short then + local suffix = first.suffix + if suffix then + ctx_btxsetfirst(short .. suffix) + else + ctx_btxsetfirst(short) + end + return true end - return true end - end - function citevariants.short(presets) - processcite(presets,{ - compress = false, - -- variant = presets.variant or "short", - setter = setter, - getter = getter, - }) - end + function citevariants.short(presets) + processcite(presets,{ + compress = false, + -- variant = presets.variant or "short", + setter = setter, + getter = getter, + }) + end -end + end --- pages (no compress) + -- pages (no compress) -do + do - local function setter(dataset,tag,entry,internal) - return { - dataset = dataset, - tag = tag, - internal = internal, - pages = getdetail(dataset,tag,"pages"), - } - end + local function setter(dataset,tag,entry,internal) + return { + dataset = dataset, + tag = tag, + internal = internal, + pages = getdetail(dataset,tag,"pages"), + } + end - local function getter(first,last) - local pages = first.pages - if pages then - if type(pages) == "table" then - ctx_btxsetfirst(pages[1]) - ctx_btxsetsecond(pages[2]) - else - ctx_btxsetfirst(pages) + local function getter(first,last) + local pages = first.pages + if pages then + if type(pages) == "table" then + ctx_btxsetfirst(pages[1]) + ctx_btxsetsecond(pages[2]) + else + ctx_btxsetfirst(pages) + end + return true end - return true end - end - function citevariants.page(presets) - processcite(presets,{ - -- variant = presets.variant or "page", - setter = setter, - getter = getter, - }) + function citevariants.page(presets) + processcite(presets,{ + -- variant = presets.variant or "page", + setter = setter, + getter = getter, + }) + end + end -end + -- num --- num + do -do + local function setter(dataset,tag,entry,internal) + local entries = entry.entries + local text = entries and entries.text or "?" + return { + tag = tag, + internal = internal, + num = text, + sortkey = text, + } + end - local function setter(dataset,tag,entry,internal) - local entries = entry.entries - local text = entries and entries.text or "?" - return { - tag = tag, - internal = internal, - num = text, - sortkey = text, - } - end + local function getter(first,last) + return simplegetter(first,last,"num") + end - local function getter(first,last) - return simplegetter(first,last,"num") - end + function citevariants.num(presets) + processcite(presets,{ + -- variant = presets.variant or "num", + setter = setter, + getter = getter, + }) + end - function citevariants.num(presets) - processcite(presets,{ - -- variant = presets.variant or "num", - setter = setter, - getter = getter, - }) end -end + -- year --- year + do -do + local function setter(dataset,tag,entry,internal) + return { + dataset = dataset, + tag = tag, + internal = internal, + year = getfield(dataset,tag,"year"), + suffix = getdetail(dataset,tag,"suffix"), + sortkey = getdetail(dataset,tag,"suffixedyear"), + } + end - local function setter(dataset,tag,entry,internal) - return { - dataset = dataset, - tag = tag, - internal = internal, - year = getfield(dataset,tag,"year"), - suffix = getdetail(dataset,tag,"suffix"), - sortkey = getdetail(dataset,tag,"suffixedyear"), - } - end + local function getter(first,last) + return simplegetter(first,last,"year") + end - local function getter(first,last) - return simplegetter(first,last,"year") - end + function citevariants.year(presets) + processcite(presets,{ + -- variant = presets.variant or "year", + setter = setter, + getter = getter, + }) + end - function citevariants.year(presets) - processcite(presets,{ - -- variant = presets.variant or "year", - setter = setter, - getter = getter, - }) end -end + -- index | serial --- index | serial + do -do + local function setter(dataset,tag,entry,internal) + local index = getfield(dataset,tag,"index") + return { + dataset = dataset, + tag = tag, + internal = internal, + index = index, + sortkey = index, + } + end - local function setter(dataset,tag,entry,internal) - local index = getfield(dataset,tag,"index") - return { - dataset = dataset, - tag = tag, - internal = internal, - index = index, - sortkey = index, - } - end + local function getter(first,last) + return simplegetter(first,last,"index") + end - local function getter(first,last) - return simplegetter(first,last,"index") - end + function citevariants.index(presets) + processcite(presets,{ + -- variant = presets.variant or "index", + setter = setter, + getter = getter, + }) + end - function citevariants.index(presets) - processcite(presets,{ - -- variant = presets.variant or "index", - setter = setter, - getter = getter, - }) - end + function citevariants.serial(presets) + processcite(presets,{ + -- variant = presets.variant or "serial", + setter = setter, + getter = getter, + }) + end - function citevariants.serial(presets) - processcite(presets,{ - -- variant = presets.variant or "serial", - setter = setter, - getter = getter, - }) end -end + -- category | type --- category | type + do -do + local function setter(dataset,tag,entry,internal) + return { + dataset = dataset, + tag = tag, + internal = internal, + category = getfield(dataset,tag,"category"), + } + end - local function setter(dataset,tag,entry,internal) - return { - dataset = dataset, - tag = tag, - internal = internal, - category = getfield(dataset,tag,"category"), - } - end + local function getter(first,last) + return simplegetter(first,last,"category") + end - local function getter(first,last) - return simplegetter(first,last,"category") - end + function citevariants.category(presets) + processcite(presets,{ + -- variant = presets.variant or "serial", + setter = setter, + getter = getter, + }) + end - function citevariants.category(presets) - processcite(presets,{ - -- variant = presets.variant or "serial", - setter = setter, - getter = getter, - }) - end + function citevariants.type(presets) + processcite(presets,{ + -- variant = presets.variant or "type", + setter = setter, + getter = getter, + }) + end - function citevariants.type(presets) - processcite(presets,{ - -- variant = presets.variant or "type", - setter = setter, - getter = getter, - }) end -end + -- key | tag --- key | tag + do -do + local function setter(dataset,tag,entry,internal) + return { + dataset = dataset, + tag = tag, + internal = internal, + } + end - local function setter(dataset,tag,entry,internal) - return { - dataset = dataset, - tag = tag, - internal = internal, - } - end + local function getter(first,last) + ctx_btxsetfirst(first.tag) + return true + end - local function getter(first,last) - ctx_btxsetfirst(first.tag) - return true - end + function citevariants.key(presets) + return processcite(presets,{ + variant = "key", + setter = setter, + getter = getter, + }) + end - function citevariants.key(presets) - return processcite(presets,{ - variant = "key", - setter = setter, - getter = getter, - }) - end + function citevariants.tag(presets) + return processcite(presets,{ + variant = "tag", + setter = setter, + getter = getter, + }) + end - function citevariants.tag(presets) - return processcite(presets,{ - variant = "tag", - setter = setter, - getter = getter, - }) end -end + -- todo : sort + -- todo : choose between publications or commands namespace + -- todo : use details.author + -- todo : sort details.author + -- (name, name and name) .. how names? how sorted? + -- todo: we loop at the tex end .. why not here + -- \cite[{hh,afo},kvm] --- todo : sort --- todo : choose between publications or commands namespace --- todo : use details.author --- todo : sort details.author --- (name, name and name) .. how names? how sorted? --- todo: we loop at the tex end .. why not here --- \cite[{hh,afo},kvm] + -- common --- common + do -do + local getauthor = publications.authors.getauthor - local getauthor = publications.authors.getauthor - - local currentbtxciteauthor = function() - context.currentbtxciteauthor() - return true -- needed? - end + local currentbtxciteauthor = function() + context.currentbtxciteauthor() + return true -- needed? + end - local function authorcompressor(found) - local result = { } - local entries = { } - for i=1,#found do - local entry = found[i] - local author = entry.author - if author then - local aentries = entries[author] - if aentries then - aentries[#aentries+1] = entry - else - entries[author] = { entry } + local function authorcompressor(found) + local result = { } + local entries = { } + for i=1,#found do + local entry = found[i] + local author = entry.author + if author then + local aentries = entries[author] + if aentries then + aentries[#aentries+1] = entry + else + entries[author] = { entry } + end end end - end - for i=1,#found do - local entry = found[i] - local author = entry.author - if author then - local aentries = entries[author] - if not aentries then - result[#result+1] = entry - elseif aentries == true then - -- already done - else - result[#result+1] = entry - entry.entries = aentries - entries[author] = true + for i=1,#found do + local entry = found[i] + local author = entry.author + if author then + local aentries = entries[author] + if not aentries then + result[#result+1] = entry + elseif aentries == true then + -- already done + else + result[#result+1] = entry + entry.entries = aentries + entries[author] = true + end end end + -- todo: add letters (should we then tag all?) + return result end - -- todo: add letters (should we then tag all?) - return result - end - local function authorconcat(target,key,setup) - ctx_btxstartsubcite(setup) - local nofcollected = #target - if nofcollected == 0 then - unknowncite(tag) - else - for i=1,nofcollected do - local entry = target[i] - local first = entry.first - local tag = entry.tag - local currentcitation = markcite(entry.dataset,tag) - ctx_btxstartciteauthor() - ctx_btxsettag(tag) - ctx_btxsetbacklink(currentcitation) - local bl = listtocite[currentcitation] - ctx_btxsetinternal(bl and bl.references.internal or "") - if first then - ctx_btxsetfirst(first[key] or f_missing(first.tag)) - local suffix = entry.suffix - local value = entry.last[key] - if value then - ctx_btxsetsecond(value) - end - if suffix then - ctx_btxsetthird(suffix) - end - else - local suffix = entry.suffix - local value = entry[key] or f_missing(tag) - ctx_btxsetfirst(value) - if suffix then - ctx_btxsetthird(suffix) + local function authorconcat(target,key,setup) + ctx_btxstartsubcite(setup) + local nofcollected = #target + if nofcollected == 0 then + unknowncite(tag) + else + for i=1,nofcollected do + local entry = target[i] + local first = entry.first + local tag = entry.tag + local currentcitation = markcite(entry.dataset,tag) + ctx_btxstartciteauthor() + ctx_btxsettag(tag) + ctx_btxsetbacklink(currentcitation) + local bl = listtocite[currentcitation] + ctx_btxsetinternal(bl and bl.references.internal or "") + if first then + ctx_btxsetfirst(first[key] or f_missing(first.tag)) + local suffix = entry.suffix + local value = entry.last[key] + if value then + ctx_btxsetsecond(value) + end + if suffix then + ctx_btxsetthird(suffix) + end + else + local suffix = entry.suffix + local value = entry[key] or f_missing(tag) + ctx_btxsetfirst(value) + if suffix then + ctx_btxsetthird(suffix) + end end + ctx_btxsetconcat(concatstate(i,nofcollected)) + ctx_btxcitesetup(setup) + ctx_btxstopciteauthor() end - ctx_btxsetconcat(concatstate(i,nofcollected)) - ctx_btxcitesetup(setup) - ctx_btxstopciteauthor() - end - end - ctx_btxstopsubcite() - end - - local function authorsingle(entry,key,setup) - ctx_btxstartsubcite(setup) - ctx_btxstartciteauthor() - local tag = entry.tag - ctx_btxsettag(tag) - -- local currentcitation = markcite(entry.dataset,tag) - -- ctx_btxsetbacklink(currentcitation) - -- local bl = listtocite[currentcitation] - -- ctx_btxsetinternal(bl and bl.references.internal or "") - ctx_btxsetfirst(entry[key] or f_missing(tag)) - ctx_btxsetthird(entry.suffix) - ctx_btxcitesetup(setup) - ctx_btxstopciteauthor() - ctx_btxstopsubcite() - end - - local partialinteractive = false - - local function authorgetter(first,last,key,setup) -- only first - -- ctx_btxsetfirst(first.author) -- unformatted - ctx_btxsetfirst(currentbtxciteauthor) -- formatter (much slower) - local entries = first.entries - -- alternatively we can use a concat with one ... so that we can only make the - -- year interactive, as with the concat - if partialinteractive and not entries then - entries = { first } - end - if entries then - local c = compresslist(entries) - local f = function() authorconcat(c,key,setup) return true end -- indeed return true? - ctx_btxsetcount(#c) - ctx_btxsetsecond(f) - else - local f = function() authorsingle(first,key,setup) return true end -- indeed return true? - ctx_btxsetcount(0) - ctx_btxsetsecond(f) + end + ctx_btxstopsubcite() end - return true - end - -- author + local function authorsingle(entry,key,setup) + ctx_btxstartsubcite(setup) + ctx_btxstartciteauthor() + local tag = entry.tag + ctx_btxsettag(tag) + -- local currentcitation = markcite(entry.dataset,tag) + -- ctx_btxsetbacklink(currentcitation) + -- local bl = listtocite[currentcitation] + -- ctx_btxsetinternal(bl and bl.references.internal or "") + ctx_btxsetfirst(entry[key] or f_missing(tag)) + ctx_btxsetthird(entry.suffix) + ctx_btxcitesetup(setup) + ctx_btxstopciteauthor() + ctx_btxstopsubcite() + end - local function setter(dataset,tag,entry,internal) - return { - dataset = dataset, - tag = tag, - internal = internal, - author = getauthor(dataset,tag,currentspecificationcategories), -- todo: list - } - end + local partialinteractive = false - local function getter(first,last,_,setup) - -- ctx_btxsetfirst(first.author) -- unformatted - ctx_btxsetfirst(currentbtxciteauthor) -- formatter (much slower) - return true - end + local function authorgetter(first,last,key,setup) -- only first + -- ctx_btxsetfirst(first.author) -- unformatted + ctx_btxsetfirst(currentbtxciteauthor) -- formatter (much slower) + local entries = first.entries + -- alternatively we can use a concat with one ... so that we can only make the + -- year interactive, as with the concat + if partialinteractive and not entries then + entries = { first } + end + if entries then + local c = compresslist(entries) + local f = function() authorconcat(c,key,setup) return true end -- indeed return true? + ctx_btxsetcount(#c) + ctx_btxsetsecond(f) + else + local f = function() authorsingle(first,key,setup) return true end -- indeed return true? + ctx_btxsetcount(0) + ctx_btxsetsecond(f) + end + return true + end - function citevariants.author(presets) - processcite(presets,{ - compress = false, - variant = "author", - setter = setter, - getter = getter, - }) - end + -- author - -- authornum + local function setter(dataset,tag,entry,internal) + return { + dataset = dataset, + tag = tag, + internal = internal, + author = getauthor(dataset,tag,currentspecification.categories), -- todo: list + } + end - local function setter(dataset,tag,entry,internal) - local text = entry.entries.text - return { - dataset = dataset, - tag = tag, - internal = internal, - author = getauthor(dataset,tag,currentspecificationcategories), -- todo: list - num = text, - sortkey = text and lpegmatch(numberonly,text), - } - end + local function getter(first,last,_,setup) + -- ctx_btxsetfirst(first.author) -- unformatted + ctx_btxsetfirst(currentbtxciteauthor) -- formatter (much slower) + return true + end - local function getter(first,last) - authorgetter(first,last,"num","author:num") - return true - end + function citevariants.author(presets) + processcite(presets,{ + compress = false, + variant = "author", + setter = setter, + getter = getter, + }) + end - function citevariants.authornum(presets) - processcite(presets,{ - variant = "authornum", - setter = setter, - getter = getter, - compressor = authorcompressor, - }) - end + -- authornum - -- authoryear | authoryears + local function setter(dataset,tag,entry,internal) + local text = entry.entries.text + return { + dataset = dataset, + tag = tag, + internal = internal, + author = getauthor(dataset,tag,currentspecification.categories), -- todo: list + num = text, + sortkey = text and lpegmatch(numberonly,text), + } + end - local function setter(dataset,tag,entry,internal) - return { - dataset = dataset, - tag = tag, - internal = internal, - author = getauthor(dataset,tag,currentspecificationcategories), -- todo: list - year = getfield(dataset,tag,"year"), - suffix = getdetail(dataset,tag,"suffix"), - sortkey = getdetail(dataset,tag,"suffixedyear"), - } - end + local function getter(first,last) + authorgetter(first,last,"num","author:num") + return true + end - local function getter(first,last) - authorgetter(first,last,"year","author:year") - return true - end + function citevariants.authornum(presets) + processcite(presets,{ + variant = "authornum", + setter = setter, + getter = getter, + compressor = authorcompressor, + }) + end - function citevariants.authoryear(presets) - processcite(presets,{ - variant = "authoryear", - setter = setter, - getter = getter, - compressor = authorcompressor, - }) - end + -- authoryear | authoryears - local function getter(first,last) - authorgetter(first,last,"year","author:years") - return true - end + local function setter(dataset,tag,entry,internal) + return { + dataset = dataset, + tag = tag, + internal = internal, + author = getauthor(dataset,tag,currentspecification.categories), -- todo: list + year = getfield(dataset,tag,"year"), + suffix = getdetail(dataset,tag,"suffix"), + sortkey = getdetail(dataset,tag,"suffixedyear"), + } + end - function citevariants.authoryears(presets) - processcite(presets,{ - variant = "authoryears", - setter = setter, - getter = getter, - compressor = authorcompressor, - }) - end + local function getter(first,last) + authorgetter(first,last,"year","author:year") + return true + end -end + function citevariants.authoryear(presets) + processcite(presets,{ + variant = "authoryear", + setter = setter, + getter = getter, + compressor = authorcompressor, + }) + end --- List variants + local function getter(first,last) + authorgetter(first,last,"year","author:years") + return true + end -local listvariants = { } -publications.listvariants = listvariants + function citevariants.authoryears(presets) + processcite(presets,{ + variant = "authoryears", + setter = setter, + getter = getter, + compressor = authorcompressor, + }) + end -function commands.btxlistvariant(dataset,block,tag,variant,listindex) - local action = listvariants[variant] or listvariants.default - if action then - action(dataset,block,tag,variant,tonumber(listindex) or 0) end -end -function listvariants.default(dataset,block,tag,variant) - ctx_btxsetfirst("?") - ctx_btxlistsetup(variant) end -function listvariants.num(dataset,block,tag,variant,listindex) - ctx_btxsetfirst(listindex) - ctx_btxlistsetup(variant) -end +-- List variants + +do -listvariants[v_yes] = listvariants.num -listvariants.bib = listvariants.num + local listvariants = { } + publications.listvariants = listvariants -function listvariants.short(dataset,block,tag,variant,listindex) - local short = getdetail(dataset,tag,"short","short") - local suffix = getdetail(dataset,tag,"suffix","suffix") - if short then - ctx_btxsetfirst(short) + function commands.btxlistvariant(dataset,block,tag,variant,listindex) + local action = listvariants[variant] or listvariants.default + if action then + action(dataset,block,tag,variant,tonumber(listindex) or 0) + end end - if suffix then - ctx_btxsetthird(suffix) + + function listvariants.default(dataset,block,tag,variant) + ctx_btxsetfirst("?") + ctx_btxlistsetup(variant) + end + + function listvariants.num(dataset,block,tag,variant,listindex) + ctx_btxsetfirst(listindex) + ctx_btxlistsetup(variant) end - ctx_btxlistsetup(variant) -end -function listvariants.page(dataset,block,tag,variant,listindex) - local rendering = renderings[dataset] - local specification = rendering.list[listindex] - for i=3,#specification do - local backlink = tonumber(specification[i]) - if backlink then - local citation = citetolist[backlink] - if citation then - local references = citation.references - if references then - local internal = references.internal - local realpage = references.realpage - if internal and realpage then - ctx_btxsetconcat(i-2) - ctx_btxsetfirst(realpage) - ctx_btxsetsecond(backlink) - ctx_btxlistsetup(variant) + listvariants[v_yes] = listvariants.num + listvariants.bib = listvariants.num + + function listvariants.short(dataset,block,tag,variant,listindex) + local short = getdetail(dataset,tag,"short","short") + local suffix = getdetail(dataset,tag,"suffix","suffix") + if short then + ctx_btxsetfirst(short) + end + if suffix then + ctx_btxsetthird(suffix) + end + ctx_btxlistsetup(variant) + end + + function listvariants.page(dataset,block,tag,variant,listindex) + local rendering = renderings[dataset] + local specification = rendering.list[listindex] + for i=3,#specification do + local backlink = tonumber(specification[i]) + if backlink then + local citation = citetolist[backlink] + if citation then + local references = citation.references + if references then + local internal = references.internal + local realpage = references.realpage + if internal and realpage then + ctx_btxsetconcat(i-2) + ctx_btxsetfirst(realpage) + ctx_btxsetsecond(backlink) + ctx_btxlistsetup(variant) + end end end end end end + end diff --git a/tex/context/base/publ-ini.mkiv b/tex/context/base/publ-ini.mkiv index f2a45584b..2c81f294e 100644 --- a/tex/context/base/publ-ini.mkiv +++ b/tex/context/base/publ-ini.mkiv @@ -65,6 +65,7 @@ \def\s!btxrtx {btxrtx} \def\s!btxatx {btxatx} \def\s!btxbtx {btxbtx} +\def\s!btxspc {btxspc} \definelabelclass[btxlabel][2] @@ -309,6 +310,58 @@ \installcommandhandler \??btxregister {btxregister} \??btxregister \installcommandhandler \??btxrendering {btxrendering} \??btxrendering +%D The following two helpers permits us to use prefixes (if we want): +%D +%D \startbuffer +%D \let\btxciteparameter\btxspecificationciteparameter +%D \let\btxlistparameter\btxspecificationlistparameter +%D +%D \edef\currentbtxspecification {apa} +%D \edef\currentbtxcitealternative{author} +%D +%D \setupbtxcitevariant [crap=crap] +%D \definebtxcitevariant [author] [check=author,extra=author] +%D \definebtxcitevariant [authoryear] [author] [check=authoryear] +%D \definebtxcitevariant [authoryears] [authoryear] [check=authoryears] +%D \setupbtxcitevariant [author] [apa:check=apa-author] +%D \setupbtxcitevariant [authoryear] [apa:check=apa-authoryear] +%D +%D \starttabulate[|lT|lT|] +%D \NC \bf check \EQ \btxciteparameter{check} \NC\NR +%D \NC \bf extra \EQ \btxciteparameter{extra} \NC\NR +%D \NC \bf crap \EQ \btxciteparameter{crap} \NC\NR +%D \stoptabulate +%D \stopbuffer +%D +%D \typebuffer \start \getbuffer \stop + +\def\btxspecificationciteparameter#1% + {\csname + \??btxcitevariant + \ifcsname\??btxcitevariant\currentbtxcitealternative:\currentbtxspecification:#1\endcsname + \currentbtxcitealternative:\currentbtxspecification:#1% + \else\ifcsname\??btxcitevariant\currentbtxcitealternative:#1\endcsname + \currentbtxcitealternative:#1% + \else + :#1% we assume defined variants + \fi\fi + \endcsname} + +\def\btxspecificationlistparameter#1% + {\csname + \??btxlistvariant + \ifcsname\??btxlistvariant\currentbtxlistalternative:\currentbtxspecification:#1\endcsname + \currentbtxlistalternative:\currentbtxspecification:#1% + \else\ifcsname\??btxlistvariant\currentbtxlistalternative:#1\endcsname + \currentbtxlistalternative:#1% + \else + :#1% we assume defined variants + \fi\fi + \endcsname} + +% \let\btxciteparameter\btxspecificationciteparameter +% \let\btxlistparameter\btxspecificationlistparameter + \appendtoks \ifx\currentbtxlistvariant\empty \else \ctxcommand{registerbtxlistvariant("\currentbtxlistvariant","\currentbtxlistvariantparent")}% @@ -322,14 +375,25 @@ \to \everydefinebtxcitevariant \unexpanded\def\usebtxdataset - {\dodoubleargument\publ_use_dataset} + {\begingroup + \dotripleempty\publ_use_dataset} -\def\publ_use_dataset[#1][#2]% - {\ifsecondargument - \ctxcommand{usebtxdataset("#1","#2")}% - \else - \ctxcommand{usebtxdataset("\v!standard","#1")}% - \fi} +\def\publ_use_dataset[#1][#2][#3]% + {\getdummyparameters[\c!specification=\currentbtxspecification,#3]% + \ifsecondargument + \ctxcommand{usebtxdataset{ + specification = "\dummyparameter\c!specification", + dataset = "#1", + filename = "#2", + }}% + \else\iffirstargument + \ctxcommand{usebtxdataset{ + specification = "\dummyparameter\c!specification", + dataset = "\v!standard", + filename = "#1", + }}% + \fi\fi + \endgroup} \definebtxdataset [\v!standard] @@ -862,6 +926,7 @@ \strc_references_direct_full_user {\s!btxset="\currentbtxdataset",% \s!btxref="\currentbtxtag",% + \s!btxspc="\currentbtxspecification",% \ifx\currentbtxbefore\empty\else\s!btxbtx={\currentbtxbefore},\fi% \ifx\currentbtxafter \empty\else\s!btxatx={\currentbtxafter },\fi% \ifx\currentbtxbacklink\currentbtxbacktrace\s!btxint="\currentbtxbacklink"\else\s!btxbck="\currentbtxbacktrace"\fi}% @@ -1351,13 +1416,14 @@ \unexpanded\def\publ_registers_set {\ifx\currentbtxregister\empty \else \ctxcommand{setbtxregister { - name = "\currentbtxregister", - state = "\btxregisterparameter\c!state", - dataset = "\btxregisterparameter\c!dataset", - field = "\btxregisterparameter\c!field", - register = "\btxregisterparameter\c!register", - method = "\btxregisterparameter\c!method", - alternative = "\btxregisterparameter\c!alternative", + specification = "\currentbtxspecification", + name = "\currentbtxregister", + state = "\btxregisterparameter\c!state", + dataset = "\btxregisterparameter\c!dataset", + field = "\btxregisterparameter\c!field", + register = "\btxregisterparameter\c!register", + method = "\btxregisterparameter\c!method", + alternative = "\btxregisterparameter\c!alternative", }}% \fi} diff --git a/tex/context/base/publ-jrn.lua b/tex/context/base/publ-jrn.lua index de98608c7..91df21904 100644 --- a/tex/context/base/publ-jrn.lua +++ b/tex/context/base/publ-jrn.lua @@ -16,30 +16,31 @@ if not modules then modules = { } end modules ['publ-jrn'] = { -- Abhandlungen der Naturforschenden Gesellschaft in Zürich = Abh. Nat.forsch. Ges. Zür. -- Abhandlungen des Naturwissenschaftlichen Vereins zu Bremen = Abh. Nat.wiss. Ver. Bremen -if not characters then require("char-utf") end +local context = context +local commands = commands local find = string.find local P, C, S, Cs, lpegmatch, lpegpatterns = lpeg.P, lpeg.C, lpeg.S, lpeg.Cs, lpeg.match, lpeg.patterns -local lower = characters.lower - local report_journals = logs.reporter("publications","journals") -publications = publications or { } +local publications = publications local journals = { } publications.journals = journals +local lowercase = characters.lower + local expansions = { } local abbreviations = { } local nofexpansions = 0 local nofabbreviations = 0 -local valid = 1 - S([[ ."':;,-]]) -local pattern = Cs((valid^1 + P(1)/"")^1) +local valid = 1 - S([[ ."':;,-]]) +local pattern = Cs((valid^1 + P(1)/"")^1) local function simplify(name) -- we have utf but it doesn't matter much if we lower the bytes - return name and lower(lpegmatch(pattern,name)) or name + return name and lowercase(lpegmatch(pattern,name)) or name end local function add(expansion,abbreviation) diff --git a/tex/context/base/publ-oth.lua b/tex/context/base/publ-oth.lua index ff7e0e29c..55c62c31e 100644 --- a/tex/context/base/publ-oth.lua +++ b/tex/context/base/publ-oth.lua @@ -11,8 +11,10 @@ local lpegmatch = lpeg.match local p_endofline = lpeg.patterns.newline -local loaders = publications.loaders -local getindex = publications.getindex +local publications = publications + +local loaders = publications.loaders +local getindex = publications.getindex local function addfield(t,k,v,fields) k = fields[k] diff --git a/tex/context/base/publ-reg.lua b/tex/context/base/publ-reg.lua index 7ff284d49..ef4ed06a3 100644 --- a/tex/context/base/publ-reg.lua +++ b/tex/context/base/publ-reg.lua @@ -21,10 +21,14 @@ local v_standard = variables.standard local v_stop = variables.stop local v_all = variables.all +local publications = publications local datasets = publications.datasets -local specifications = { } +local specifications = publications.specifications +local detailed = publications.detailed + +local registrations = { } local sequence = { } -local flushers = table.setmetatableindex(function(t,k) t[k] = default return default end) +local flushers = table.setmetatableindex(function(t,k) local v = t.default t[k] = v return v end) function commands.setbtxregister(specification) local name = specification.name @@ -38,10 +42,10 @@ function commands.setbtxregister(specification) dataset = v_all end -- could be metatable magic - local s = specifications[register] + local s = registrations[register] if not s then s = { } - specifications[register] = s + registrations[register] = s end local d = s[dataset] if not d then @@ -51,17 +55,24 @@ function commands.setbtxregister(specification) -- -- check all -- + local processors = name ~= register and name or "" + if processor == "" then + processor = nil + elseif processor then + processor = "btx:r:" .. processor + end + -- d.active = specification.state ~= v_stop d.once = specification.method == v_once or false d.field = field - d.processor = name ~= register and name or "" + d.processor = processor d.alternative = d.alternative or specification.alternative d.register = register d.dataset = dataset d.done = d.done or { } -- sequence = { } - for register, s in sortedhash(specifications) do + for register, s in sortedhash(registrations) do for dataset, d in sortedhash(s) do if d.active then sequence[#sequence+1] = d @@ -70,21 +81,50 @@ function commands.setbtxregister(specification) end end +----- getter = publications.directget + +local function getter(current,tag,step) -- todo: detail + local data = current.luadata[tag] + if data then + local catspec = specifications[step.specification].categories[data.category] + if catspec then + local fields = catspec.fields + if fields then + local field = step.field + local sets = catspec.sets + if sets then + local set = sets[field] + if set then + for i=1,#set do + local field = set[i] + local value = fields[field] and data[field] -- redundant check + if value then + return field, value, catspec.types[field] or "string" + end + end + end + end + local value = fields[field] and data[field] + if value then + return field, value, catspec.types[field] or "string" + end + end + end + end +end + function commands.btxtoregister(dataset,tag) + local current = datasets[dataset] for i=1,#sequence do local step = sequence[i] local dset = step.dataset if dset == v_all or dset == dataset then local done = step.done if not done[tag] then - local current = datasets[dataset] - local entry = current.luadata[tag] - if entry then - local processor = step.processor - if processor and processor ~= "" then - step.processor = "btx:r:" .. processor - end - flushers[step.field or "default"](step,dataset,tag,current,entry,current.details[tag]) + local field, value, kind = getter(current,tag,step) + if value then + local cast = detailed[kind][value] or value + flushers[kind](step,field,value,cast) end done[tag] = true end @@ -102,53 +142,44 @@ end local ctx_dosetfastregisterentry = context.dosetfastregisterentry -- register entry key -local p_keywords = lpeg.tsplitat(lpeg.patterns.whitespace^0 * lpeg.P(";") * lpeg.patterns.whitespace^0) -local serialize = publications.serializeauthor -local components = publications.authorcomponents -local f_author = formatters[ [[\btxindexedauthor{%s}{%s}{%s}{%s}{%s}{%s}]] ] +local p_keywords = lpeg.tsplitat(lpeg.patterns.whitespace^0 * lpeg.P(";") * lpeg.patterns.whitespace^0) +local serialize = publications.serializeauthor +local components = publications.authorcomponents +local f_author = formatters[ [[\btxindexedauthor{%s}{%s}{%s}{%s}{%s}{%s}]] ] -function flushers.default(specification,dataset,tag,current,entry,detail) - local field = specification.field - local k = detail[field] or entry[field] - if k then - ctx_dosetfastregisterentry(specification.register,k,"",specification.processor,"") +function flushers.string(step,field,value,cast) + if value and value ~= "" then + ctx_dosetfastregisterentry(step.register,type(cast) == "string" and cast or value,"",step.processor or "","") end end +flushers.default = flushers.string + local shorts = { - normalshort = true, - invertedshort = true, + normalshort = "normalshort", + invertedshort = "invertedshort", } -function flushers.author(specification,dataset,tag,current,entry,detail) - if detail then - local field = specification.field - local author = detail[field] - if author then - local alternative = specification.alternative or "invertedshort" - local short = shorts[alternative] - local register = specification.register - local processor = specification.processor - for i=1,#author do - local a = author[i] - local k = serialize(a) - local e = f_author(alternative,components(a,short)) - ctx_dosetfastregisterentry(register,e,k,processor,"") - end +function flushers.author(step,field,value,cast) + if cast and #cast > 0 then + local register = step.register + local processor = step.processor + local alternative = shorts[step.alternative or "invertedshort"] or "invertedshort" + for i=1,#cast do + local a = cast[i] + local k = serialize(a) + local e = f_author(alternative,components(a,short)) + ctx_dosetfastregisterentry(register,e,k,processor or "","") end end end -function flushers.keywords(specification,dataset,tag,current,entry,detail) - if entry then - local value = entry[specification.field] - if value then - local keywords = lpegmatch(p_keywords,value) - local register = specification.register - local processor = specification.processor - for i=1,#keywords do - ctx_dosetfastregisterentry(register,keywords[i],"",processor,"") - end +function flushers.keywords(step,field,value,cast) + if cast and #cast > 0 then + local register = step.register + local processor = step.processor + for i=1,#cast do + ctx_dosetfastregisterentry(register,cast[i],"",processor or "","") end end end diff --git a/tex/context/base/publ-tra.lua b/tex/context/base/publ-tra.lua index 09dab888a..d3bc05ea1 100644 --- a/tex/context/base/publ-tra.lua +++ b/tex/context/base/publ-tra.lua @@ -12,18 +12,21 @@ local sortedhash, sortedkeys = table.sortedhash, table.sortedkeys local settings_to_array = utilities.parsers.settings_to_array local formatters = string.formatters -local tracers = publications.tracers or { } +local context = context +local commands = commands + +local publications = publications +local tracers = publications.tracers +local tables = publications.tables local datasets = publications.datasets local specifications = publications.specifications -local context = context - local ctx_NC, ctx_NR, ctx_HL, ctx_FL, ctx_ML, ctx_LL = context.NC, context.NR, context.HL, context.FL, context.ML, context.LL local ctx_bold, ctx_monobold, ctx_rotate, ctx_llap, ctx_rlap = context.bold, context.formatted.monobold, context.rotate, context.llap, context.rlap local ctx_starttabulate, ctx_stoptabulate = context.starttabulate, context.stoptabulate -local privates = publications.tables.privates -local specials = publications.tables.specials +local privates = tables.privates +local specials = tables.specials local report = logs.reporter("publications","tracers") @@ -35,7 +38,7 @@ function tracers.showdatasetfields(settings) local fielddata = specification and specifications[specification] or specifications.apa local categories = fielddata.categories if next(luadata) then - ctx_starttabulate { "|lT|lT|pT|" } + ctx_starttabulate { "|lT|lT|pTl|" } ctx_NC() ctx_bold("tag") ctx_NC() ctx_bold("category") ctx_NC() ctx_bold("fields") @@ -58,9 +61,9 @@ function tracers.showdatasetfields(settings) if kind == "required" then context("{\\darkgreen %s} ",key) elseif kind == "optional" then - context("{\\darkyellow %s} ",key) - else context("%s ",key) + else + context("{\\darkyellow %s} ",key) end end end @@ -81,7 +84,7 @@ function tracers.showdatasetcompleteness(settings) local lpegmatch = lpeg.match local texescape = lpeg.patterns.texescape - local preamble = { "|lTBw(5em)|lBTp(10em)|p|" } + local preamble = { "|lTBw(5em)|lBTp(10em)|pl|" } local function identified(tag,category,crossref) ctx_NC() @@ -284,7 +287,7 @@ function tracers.showfields(settings) end function tracers.showtables(settings) - for name, list in sortedhash(publications.tables) do + for name, list in sortedhash(tables) do ctx_starttabulate { "|Tl|Tl|" } ctx_FL() ctx_NC() diff --git a/tex/context/base/publ-usr.lua b/tex/context/base/publ-usr.lua index e2100a289..4650401ee 100644 --- a/tex/context/base/publ-usr.lua +++ b/tex/context/base/publ-usr.lua @@ -7,9 +7,11 @@ if not modules then modules = { } end modules ['publ-usr'] = { } local P, Cs, R, Cc, Carg = lpeg.P, lpeg.Cs, lpeg.R, lpeg.Cc, lpeg.Carg - +local lpegmatch = lpeg.match local settings_to_hash = utilities.parsers.settings_to_hash --- local chardata = characters.data + +local publications = publications +local datasets = publications.datasets -- local str = [[ -- \startpublication[k=Berdnikov:TB21-2-129,t=article,a={{Berdnikov},{}},y=2000,n=2257,s=BHHJ00] @@ -70,8 +72,10 @@ local leftbrace = P("{") local rightbrace = P("}") local leftbracket = P("[") local rightbracket = P("]") +local backslash = P("\\") +local letter = R("az","AZ") -local key = P("\\") * Cs(R("az","AZ")^1) * lpeg.patterns.space^0 +local key = backslash * Cs(letter^1) * lpeg.patterns.space^0 local mandate = leftbrace * Cs(lpeg.patterns.balanced) * rightbrace + Cc(false) local optional = leftbracket * Cs((1-rightbracket)^0) * rightbracket + Cc(false) local value = optional^-1 * mandate^-1 * optional^-1 * mandate^-2 @@ -79,14 +83,17 @@ local value = optional^-1 * mandate^-1 * optional^-1 * mandate^-2 local pattern = ((Carg(1) * key * value) / register + P(1))^0 function publications.addtexentry(dataset,settings,content) - settings = settings_to_hash(settings) + local current = datasets[dataset] + local settings = settings_to_hash(settings) local data = { tag = settings.tag or settings.k or "no tag", category = settings.category or settings.t or "article", } - lpeg.match(pattern,content,1,data) -- can set tag too - dataset.userdata[data.tag] = data - dataset.luadata[data.tag] = data - publications.markasupdated(dataset) + lpegmatch(pattern,content,1,data) -- can set tag too + current.userdata[data.tag] = data + current.luadata[data.tag] = data + publications.markasupdated(current) return data end + +commands.addbtxentry = publications.addtexentry diff --git a/tex/context/base/status-files.pdf b/tex/context/base/status-files.pdf Binary files differindex 0cd6b92cb..4f2bb757c 100644 --- a/tex/context/base/status-files.pdf +++ b/tex/context/base/status-files.pdf diff --git a/tex/context/base/status-lua.pdf b/tex/context/base/status-lua.pdf Binary files differindex 5182fb633..c08793164 100644 --- a/tex/context/base/status-lua.pdf +++ b/tex/context/base/status-lua.pdf diff --git a/tex/context/base/status-mkiv.lua b/tex/context/base/status-mkiv.lua index 3c1866f6f..65df5dd40 100644 --- a/tex/context/base/status-mkiv.lua +++ b/tex/context/base/status-mkiv.lua @@ -2572,6 +2572,12 @@ return { }, { category = "mkiv", + filename = "publ-jrn", + loading = "always", + status = "pending", + }, + { + category = "mkiv", filename = "publ-xml", loading = "always", status = "pending", diff --git a/tex/generic/context/luatex/luatex-fonts-merged.lua b/tex/generic/context/luatex/luatex-fonts-merged.lua index dc2bb35bf..317ab6d92 100644 --- a/tex/generic/context/luatex/luatex-fonts-merged.lua +++ b/tex/generic/context/luatex/luatex-fonts-merged.lua @@ -1,6 +1,6 @@ -- merged file : luatex-fonts-merged.lua -- parent file : luatex-fonts.lua --- merge date : 11/11/14 12:12:33 +-- merge date : 11/12/14 11:56:31 do -- begin closure to overcome local limits and interference |