Add local caching of "simple" Graphics State (ExtGState) data in PartialEvaluator.{getOperatorList, getTextContent} (issue 2813)
#12087
Conversation
From: Bot.io (Linux m4)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.67.70.0:8877/de2b47a8d00ce69/output.txt |
From: Bot.io (Windows)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.215.176.217:8877/5e608eacd2ef0bd/output.txt |
From: Bot.io (Linux m4)FailedFull output at http://54.67.70.0:8877/de2b47a8d00ce69/output.txt Total script time: 26.66 mins
Image differences available at: http://54.67.70.0:8877/de2b47a8d00ce69/reftest-analyzer.html#web=eq.log |
From: Bot.io (Windows)FailedFull output at http://54.215.176.217:8877/5e608eacd2ef0bd/output.txt Total script time: 31.29 mins
Image differences available at: http://54.215.176.217:8877/5e608eacd2ef0bd/reftest-analyzer.html#web=eq.log |
Snuffleupagus
force-pushed
the
LocalGStateCache
branch
from
6ea824e to
3bc5932
Compare
Snuffleupagus
force-pushed
the
LocalGStateCache
branch
13 times, most recently
from
f2fdc0e to
7d131e6
Compare
From: Bot.io (Linux m4)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.67.70.0:8877/83903137cb25f06/output.txt |
From: Bot.io (Windows)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.215.176.217:8877/30816e7aa97ed50/output.txt |
From: Bot.io (Linux m4)FailedFull output at http://54.67.70.0:8877/83903137cb25f06/output.txt Total script time: 26.65 mins
Image differences available at: http://54.67.70.0:8877/83903137cb25f06/reftest-analyzer.html#web=eq.log |
From: Bot.io (Windows)FailedFull output at http://54.215.176.217:8877/30816e7aa97ed50/output.txt Total script time: 30.57 mins
Image differences available at: http://54.215.176.217:8877/30816e7aa97ed50/reftest-analyzer.html#web=eq.log |
|
/botio-linux preview |
From: Bot.io (Linux m4)ReceivedCommand cmd_preview from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.67.70.0:8877/b82ead3d1bc23ec/output.txt |
From: Bot.io (Linux m4)SuccessFull output at http://54.67.70.0:8877/b82ead3d1bc23ec/output.txt Total script time: 3.36 mins Published |
Snuffleupagus
force-pushed
the
LocalGStateCache
branch
from
7d131e6 to
e02b693
Compare
From: Bot.io (Windows)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.215.176.217:8877/637a5dd89842027/output.txt |
From: Bot.io (Linux m4)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.67.70.0:8877/90d6977baa9680d/output.txt |
From: Bot.io (Linux m4)FailedFull output at http://54.67.70.0:8877/90d6977baa9680d/output.txt Total script time: 26.63 mins
Image differences available at: http://54.67.70.0:8877/90d6977baa9680d/reftest-analyzer.html#web=eq.log |
From: Bot.io (Windows)FailedFull output at http://54.215.176.217:8877/637a5dd89842027/output.txt Total script time: 29.76 mins
Image differences available at: http://54.215.176.217:8877/637a5dd89842027/reftest-analyzer.html#web=eq.log |
…tialEvaluator.getOperatorList` (issue 2813) This patch will help pathological cases the most, with issue 2813 being a particularily problematic example. While there's only *four* `/ExtGState` resources, there's a total `29062` of `setGState` operators. Even though parsing of a single `/ExtGState` resource is quite fast, having to re-parse them thousands of times does add up quite significantly. For simplicity we'll only cache "simple" `/ExtGState` resource, since e.g. the general `SMask` case cannot be easily cached (without re-factoring other code, which may have undesirable effects on general parsing). By caching "simple" `/ExtGState` resource, we thus improve performance by: - Not having to fetch/validate/parse the same `/ExtGState` data over and over. - Handling of repeated `setGState` operators becomes *synchronous* during the `OperatorList` building, instead of having to defer to the event-loop/microtask-queue since the `/ExtGState` parsing is done asynchronously. --- Obviously I had intended to include (standard) benchmark results with this patch, but for reasons I don't understand the test run-time (even with `master`) of the document in issue 2813 is *a lot* slower than in the development viewer (making normal benchmarking infeasible). However, testing this manually in the development viewer (using `pdfBug=Stats`) shows a *reduction* of `~10 %` in the rendering time of the PDF document in issue 2813.
…tialEvaluator.getTextContent` It turns out that `getTextContent` suffers from *similar* problems with repeated GStates as `getOperatorList`; please see the previous patch. While only `/ExtGState` resources containing Fonts will actually be *parsed* by `PartialEvaluator.getTextContent`, we're still forced to fetch/validate repeated `/ExtGState` resources even though *most* of them won't affect the textContent (since they mostly contain purely graphical state). With these changes we also no longer need to immediately reset the current text-state when encountering a `setGState` operator, which may thus improve text-selection in some cases.
Snuffleupagus
force-pushed
the
LocalGStateCache
branch
from
e02b693 to
981ff41
Compare
From: Bot.io (Linux m4)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.67.70.0:8877/5cb559f3e609ef5/output.txt |
From: Bot.io (Windows)ReceivedCommand cmd_test from @Snuffleupagus received. Current queue size: 0 Live output at: http://54.215.176.217:8877/39fc5ad6e7d7c5a/output.txt |
From: Bot.io (Linux m4)FailedFull output at http://54.67.70.0:8877/5cb559f3e609ef5/output.txt Total script time: 26.55 mins
Image differences available at: http://54.67.70.0:8877/5cb559f3e609ef5/reftest-analyzer.html#web=eq.log |
From: Bot.io (Windows)FailedFull output at http://54.215.176.217:8877/39fc5ad6e7d7c5a/output.txt Total script time: 30.33 mins
Image differences available at: http://54.215.176.217:8877/39fc5ad6e7d7c5a/reftest-analyzer.html#web=eq.log |
Since this method calls `Dict.get` to fetch data, there could thus be `Error`s thrown in corrupt PDF documents when attempting to resolve an indirect object. To ensure that this won't ever become a problem, we change the method to be `async` such that a rejected Promise would be returned and general OperatorList parsing won't break.
|
/botio-linux preview |
From: Bot.io (Linux m4)ReceivedCommand cmd_preview from @timvandermeij received. Current queue size: 0 Live output at: http://54.67.70.0:8877/18ab602b53471b8/output.txt |
From: Bot.io (Linux m4)SuccessFull output at http://54.67.70.0:8877/18ab602b53471b8/output.txt Total script time: 3.29 mins Published |
|
Thank you for looking into this! |
Add local caching of "simple" Graphics State (ExtGState) data in
PartialEvaluator.getOperatorList(issue 2813)This patch will help pathological cases the most, with issue Extrememly slow rendering on this pdf #2813 being a particularily problematic example. While there's only four
/ExtGStateresources, there's a total29062ofsetGStateoperators. Even though parsing of a single/ExtGStateresource is quite fast, having to re-parse them thousands of times does add up quite significantly.For simplicity we'll only cache "simple"
/ExtGStateresource, since e.g. the generalSMaskcase cannot be easily cached (without re-factoring other code, which may have undesirable effects on general parsing).By caching "simple"
/ExtGStateresource, we thus improve performance by:/ExtGStatedata over and over.setGStateoperators becomes synchronous during theOperatorListbuilding, instead of having to defer to the event-loop/microtask-queue since the/ExtGStateparsing is done asynchronously.Obviously I had intended to include (standard) benchmark results with this patch, but for reasons I don't understand the test run-time (even with
master) of the document in issue Extrememly slow rendering on this pdf #2813 is a lot slower than in the development viewer (making normal benchmarking infeasible).However, testing this manually in the development viewer (using
pdfBug=Stats) shows a reduction of~10 %in the rendering time of the PDF document in issue Extrememly slow rendering on this pdf #2813.Add local caching of non-font Graphics State (ExtGState) data in
PartialEvaluator.getTextContentIt turns out that
getTextContentsuffers from similar problems with repeated GStates asgetOperatorList; please see the previous patch.While only
/ExtGStateresources containing Fonts will actually be parsed byPartialEvaluator.getTextContent, we're still forced to fetch/validate repeated/ExtGStateresources even though most of them won't affect the textContent (since they mostly contain purely graphical state).With these changes we also no longer need to immediately reset the current text-state when encountering a
setGStateoperator, which may thus improve text-selection in some cases.