Is there a way to combine various steps into one?

emorales

Hello. Windows7. Is there a way to customize in a shortcut a series of steps I always do? Let me explain the steps; 1) I download a pdf, 2) I drag it into a folder in Zotero, 3)Retrieve Metadata For PDF, and 4)Rename File from Parent Metadata. I absolutely always do the same four steps. Is there a way to combine these four steps into one?

adamsmith

ZotFile in theory has a watch folder function to automate 1/2, though I never got that working sufficiently reliably to be worthwhile for me. I know others have, though, I think bwiernik uses it successfully.
Automating 2/3 is planned, just not implemented.
I don't think 4 is currently possible, though I could be wrong. I think that's on the table, too, though (i.e combining 3/4).

There's no way for you to customize a shortcut for those, though, short of writing a Zotero add-on, so no to your specific question.

emorales

Thanks adamsmith. Allow me related question, please. Step 2 always takes a while, I mean some good 30 seconds, with the usual comment Firefox (Not Responding) in those 30 seconds, until suddenly is done. It is kind of odd it takes so long because I have already downloaded the pdf to the Downloads folder. Is this time something rather normal?
Thanks.

adamsmith

30 secs for a single PDF? That seems way too long. What happens is that Zotero is indexing the PDF, but that should take very little time -- IIRC, when we benchmarked the indexing, all of War and Peace indexed in <10secs.
Disable all other add-ons in Firefox, see if you still get that slow performance and if so, please provide a debug ID for that process
https://www.zotero.org/support/debug_output
i.e. start the debug before you drag the item, then submit it after the freeze ends and mention how many secs you've waited.

aurimas

I've noticed that adding PDFs to my library does occasionally take some time with Zotero being unresponsive (not 30 seconds for a single PDF, but still annoyingly long ~5 seconds). Here's some debug log showing a couple significant delays in database operations.(5)(+0000000): INSERT OR IGNORE INTO fulltextWords (word) SELECT word FROM indexing.fulltextWords
(5)(+0002162): DELETE FROM fulltextItemWords WHERE itemID = ?
(5)(+0000000): Binding parameter 1 of type int: 1863
(5)(+0000020): INSERT OR IGNORE INTO fulltextItemWords (wordID, itemID) SELECT wordID, ? FROM fulltextWords JOIN indexing.fulltextWords USING(word)
(5)(+0000001): Binding parameter 1 of type int: 1863
(5)(+0004362): REPLACE INTO fulltextItems (itemID, version) VALUES (?,?)
(5)(+0000000): Binding parameter 1 of type int: 1863
(5)(+0000000): Binding parameter 2 of type int: 0
(5)(+0000008): DELETE FROM indexing.fulltextWords
I think this may occur when adding PDFs during times of higher disk activity (e.g. Zotero checking file attachments for sync purposes, which is what was happening for this debug log). In any case, I don't think there's much we can do at this point, but this should be a lot better in upcoming versions of Zotero where database access will be asynchronous.

dstillman

IIRC, when we benchmarked the indexing, all of War and Peace indexed in <10secs</blockquote>
1.5 seconds, I think, but we have had reports that it's significantly slower with large full-text indexes. Asynchronous DB access in the next major version will help, as will switching to a new full-text indexing system, which is still planned.

Gracile

we have had reports that it's significantly slower with large full-text indexes.

FWIW, I can confirm that. Around 30 seconds for a single 5-6 MB PDF.

dstillman

Gracile: Can you provide a Debug ID for that, as well as your index stats from the Search pane of the prefs? Also, values of the two indexing settings? Might not be much we can do about it at the moment, as Aurimas says, but 30 seconds seems way long, so would be good to confirm that it's the same issue.

bwiernik

@emorales

I can confirm that the ZotFile watch folder function has worked consistently on Windows for me to add PDFs and retrieve metadata. It's possible that it works less well on other systems.

Gracile

@Dan:

-Adding a PDF to Zotero, which automatically indexes it: D1256799918
-Reindexing (the same) PDF (already in Zotero): D1686451529

Max chars to index: 500000 (i.e. default)
Max pages to index: 100 (i.e. default)

Indexed: 654
Partially: 14
Not indexed: 654
Words: 190631

But I doubt that these figures are correct, it's somehow strange that my indexed and not indexed are exactly the same, 654. There are 482 PDF and 259 HTML in my library.