From: faewik(a)gmail.com Date: Mon, 9 May 2016 12:36:19 +0100 To: j.m.muller(a)hccnet.nl; glamtools(a)lists.wikimedia.org Subject: Re: [Glamtools] adding 3134 captions after upload? There's no 'painless' solution. I have made similar housekeeping updates to very large batch uploads post-upload using pywikibot using short scripts. Amongst other things, pywikibot can pull up the specific image page history and tell which past edits have been made and by whom, then make intelligent decisions on whether to add information or how to best do it. Terribly handy for avoiding doing something twice. I suggest not being in a hurry. Focus on getting the translations done, if amenable to it, it would be worth training up the translator so they write the new text directly into the image templates rather than somewhere else. If this would be too difficult, then just get the translations as a file or spreadsheet, mapped against the related Commons images or reference numbers. There are several bot writers that might be interested in plugging the extra translations seamlessly into the current image pages once the data is ready. The best place to ask for a Commons bot specialist is on the bots work requests page.[1] As you do so much of this stuff, I strongly recommend you try playing around with Python and pywikibot yourself, perhaps delaying this translation fix so you can use it to practice with. A script to do what you are looking for might only take around 20 or 30 lines of original programming, most of pywikibot's power lies in the handy modules that have been written to do exactly this sort of thing. See the advice and code examples in the manual.[2] Links 1. https://commons.wikimedia.org/wiki/Commons:Bots/Work_requests 2. https://www.mediawiki.org/wiki/Manual:Pywikibot Fae On 9 May 2016 at 12:19, Hans Muller <j.m.muller(a)hccnet.nl> wrote: > Dear All, > > > > ** Is it possible to add 3134 captions to existing images on > Commons? > ** (without removing earlier work on the images, metadata and > categories) > > This nice question presents itself in the following case. > Recently, i > uploaded 3134 images from Dutch University Library Leiden UBL and > KITLV to > > > > https://commons.wikimedia.org/wiki/Category:Media_contributed_by_ > KITL > V_2016_-_Art_-_Temp > (309 colonial works of art) > > > > and > > https://commons.wikimedia.org/wiki/Category:Media_contributed_by_ > KITL > V_2016_-_Temp > (2805 colonial photographs) > > > > with captions in Dutch and English. I corrected the metadata, > categorised the images and put them into Wikipedia articles (only > a few ;-) > > Now a translator might be paid to translate the captions into > Bahasa > Indonesia, the language of the people perhaps most interested in > these historical images. If the translator succeeds, that would > mean require language templates like > > {{id|1= ... }} > > > > to the description fields. > > Re-uploading with GWToolset proved to be impossible if another > user has worked on the metadata since upload. Categorisation and > my own corrections to the metadata would be lost. > > * Is there another, painless solution putting in the new > translation while keeping the previous work on these files? > > Thanks for you considering this problem, > Best regards from a sunny Utrecht, the Netherlands, > > > > hans muller https://commons.wikimedia.org/wiki/User:Hansmuller > -- faewik(a)gmail.com https://commons.wikimedia.org/wiki/User:Fae _______________________________________________ Glamtools mailing list Glamtools(a)lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/glamtools