Alex brollo
Iscritto il 5 giu 2008
Contenuto cancellato Contenuto aggiunto
Phe (discussione | contributi) |
|||
Riga 557:
::Io proverei lo ''splicing'', allineando perfettamente alla scansione nuova in modo che in caso di ripensamento basti caricare la nuova sulla vecchia senza spostamenti ulteriori di pagina. --[[User:Alex brollo|Alex brollo]] ([[User talk:Alex brollo|disc.]]) 16:16, 11 feb 2013 (CET)
::: Ci ho ripensato, la vecchia scansione è desolante. Penso che il ''cloning'' + spostamento di un po' di pagine senza redirect sia la soluzione migliore. --[[User:Alex brollo|Alex brollo]] ([[User talk:Alex brollo|disc.]]) 16:48, 11 feb 2013 (CET)
== hOCR ==
hi, I started to setup an hOCR server, it can retrieve data only for book I manually hocr'ized (no hocr creation on the fly nor translation of archive.org finereader xml to hocr). At the moment data are available only for one book [[Indice:Opere matematiche (Cremona) II.djvu]], here a piece of javascript to call the server [http://fr.wikisource.org/wiki/Utilisateur:Phe/hocr.js], change the ns number and you'll get an hocr tab for test purpose. Note than data['text'] is the complete hocr html but as a text string, it need to be parsed to allow iteration on element or to use jquery on it to select set of element. I'll have little time in the next few days to work on it. Quite possible I'll add more return type with a &format= parameter e.g. &format=xml to the server if serving it as text is boring to handle in javascript. If any answer, on my fr.ws page please. [[User:Phe|Phe]] ([[User talk:Phe|disc.]]) 23:13, 11 feb 2013 (CET)
| |||