Hi @Jandrews377, thanks for taking the time to report this and welcome to Wikimedia Phabricator! Which software version is this about?

Hi @Aklapper. Thanks for taking the time to reply. Our Versions:

• mediawiki 1.34
• thenets/parsoid:0.11
• pastakhov/restbase:0.16.9
• elasticsearch:6.5.4

All of the modules are using the 1.34 branch:

• Echo
• NoTitle
• Spoilers
• Thanks
• CheckUser
• Math
• ExternalData
• Flow
• UploadWizard
• Babel
• cldr
• CleanChanges
• UniversalLanguageSelector
• VisualEditor
• CirrusSearch
• Elastica
• PDFEmbed

Could not normalize image parameters for ... is the message apierror-urlparamnormal. In a quick review this looks to only be emitted by ApiQueryImageInfo.

If you try and get the image info without search, does it also error? An example api request would be something like the following, but replace the title with a local pdf. This may depend on a very specific pdf being returned, rather than any arbitrary one, hard to say.

https://commons.wikimedia.org/w/api.php?action=query&format=json&prop=imageinfo&titles=File%3AAlmanac_Kievlyanin_(book_2%2C_1841).pdf

Hi EBernhardson,

Using your query, I don't get an error.
However, using the query the gallery generates for its ajax request, I get the error:

/api.php?action=query&format=json&generator=search&gsrnamespace=6&iiurlheight=200&iiprop=dimensions%7Curl%7Cmediatype%7Cextmetadata%7Ctimestamp%7Cuser&prop=imageinfo&gsrsearch=TruTrack&iiurlwidth=300&gsroffset=0&gsrlimit=15

If I manipulate the request and remove the iiurlheight and iiurlwidth parameters, I am able to get a response, without issue.
It looks like the backend may be having trouble generating thumbnaisl for pdf's ? Unfortunately I dont see any errors.

Many thanks.

@ssastry , this looks like it may be less of a search issue and more related to the pdf handler

@MPhamWMF the PDF handler we maintain might not be related because it's a nodejs service that generates PDF on the fly for the requested wiki page and it doesn't handle uploaded PDF files.

I'm assuming this has to do with Wikibase and not the PDF handler, unless I'm missing something.

I have the exact same problem here.. Any update?

All updates are in this ticket.

Sorry, I am new here. Is there a solution to this problem?

Hi, unfortunately there isn't currently... This ticket is still open and anyone is welcome to investigate and work on a solution. :)

In my case, I removed the PdfHandler extension and it worked just fine.