Fix a PDFHistory regression with document hashes of the nameddest=... form
#9013
Conversation
…...` form Unfortunately I've just found out that this isn't working entirely correct; my apologies for accidentally breaking this in PR 8775. Compare e.g. this link: http://mirrors.ctan.org/info/lshort/english/lshort.pdf#page.157, with this one: http://mirrors.ctan.org/info/lshort/english/lshort.pdf#nameddest=page.157. Notice how in the *second* case, the history stops working correctly. *The various edge-case regressions in the new `PDFHistory` code is reminding my why I put off the rewrite for so long :-(*
Snuffleupagus
changed the title
PDFHistory regression with document hashes of nameddest=... formPDFHistory regression with document hashes of the nameddest=... form
|
/botio-linux preview |
From: Bot.io (Linux m4)ReceivedCommand cmd_preview from @timvandermeij received. Current queue size: 0 Live output at: http://54.67.70.0:8877/dc2d67dc59bad4f/output.txt |
From: Bot.io (Linux m4)SuccessFull output at http://54.67.70.0:8877/dc2d67dc59bad4f/output.txt Total script time: 2.35 mins Published |
|
/botio unittest |
From: Bot.io (Linux m4)ReceivedCommand cmd_unittest from @timvandermeij received. Current queue size: 0 Live output at: http://54.67.70.0:8877/f4c047a0a2a01ee/output.txt |
From: Bot.io (Windows)ReceivedCommand cmd_unittest from @timvandermeij received. Current queue size: 0 Live output at: http://54.215.176.217:8877/bf2cef2015a0347/output.txt |
From: Bot.io (Linux m4)SuccessFull output at http://54.67.70.0:8877/f4c047a0a2a01ee/output.txt Total script time: 2.70 mins
|
From: Bot.io (Windows)SuccessFull output at http://54.215.176.217:8877/bf2cef2015a0347/output.txt Total script time: 6.15 mins
|
|
Thank you for fixing this!
The rewrite did bring us modern and more understandable/maintainable code that is also partly unit tested. I see this as a major improvement over the old situation that had hard-to-fix issues. Big rewrites always have a higher risk for regressions, but in this case I think there are relatively few of them and they are easily fixable, so don't worry about it too much! :-) |
|
@timvandermeij As always, thanks a lot for reviewing :-) |
Fix a `PDFHistory` regression with document hashes of the `nameddest=...` form
Unfortunately I've just found out that this isn't working entirely correct; my apologies for accidentally breaking this in PR #8775.
Compare e.g. this link: http://mirrors.ctan.org/info/lshort/english/lshort.pdf#page.157, with this one: http://mirrors.ctan.org/info/lshort/english/lshort.pdf#nameddest=page.157.
Notice how in the second case, the history stops working correctly.
The various edge-case regressions in the new
PDFHistorycode is reminding my why I put off the rewrite for so long :-(