Last modified: 2013-01-15 23:48:06 UTC
You can see this template in action at http://es.wikipedia.org/w/index.php?title=Anexo:Monumentos_Hist%C3%B3ricos_de_Panam%C3%A1&action=edit§ion=7 another similar template that is not parsed correctly is Fila BIC. I'm trying to parse it with the following command cat Darien.txt | node ./js/lib/parse.js --wgScript "http://es.wikipedia.org/w" --wgScriptPath "http://es.wikipedia.org/wiki/" and I only get table headers in the output and a mention to a template but not even the name of the template.
http://parsoid.wmflabs.org/es/Anexo:Monumentos_Hist%C3%B3ricos_de_Panam%C3%A1 produces some output, but it is clearly not quite correct. I have not yet looked into the included templates, but would not be surprised if this was an instance of mis-nested parser functions as described in http://www.mediawiki.org/wiki/Parsoid/limitations.
The first noinclude end tag is tokenized as an attribute of the table tag: echo '{{Plantilla:Fila_PCN}}' | nodejs parse.js --wgScript "http://es.wikipedia.org/w" --wgScriptPath "http://es.wikipedia.org/wiki/" --debug 2>&1 | less (..) { "type": "TagTk", "name": "table", "attribs": [ { "k": "</noinclude", "v": "" } ], "dataAttribs": { "tsr": [ 11, 24 ] }, "tagToStringFns": {} }
Tokenization has since been fixed, but round-tripping is still broken: http://parsoid.wmflabs.org/_rt/es/Anexo:Monumentos_Hist%C3%B3ricos_de_Panam%C3%A1
Roundtripping fixed in https://gerrit.wikimedia.org/r/#/c/34335/
A last issue with caption attributes was fixed in https://gerrit.wikimedia.org/r/#/c/44170/. In http://parsoid.wmflabs.org/_rt/es/Anexo:Monumentos_Hist%C3%B3ricos_de_Panam%C3%A1 there are now only a few syntactic diffs (duplicate class attributes removed) left, so I would call this bug fixed.
