These changes improves the consistency ever so slightly in the `PDFOutlineViewer._dispatchEvent` method, by making sure that we can tell the following two cases apart:
- The "pagesloaded" event has *not yet* been fired.
- The "pagesloaded" event has been fired, but no pages were available.
*This patch can be tested e.g. with the `poppler-85140-0.pdf` document from the test-suite.*
For some sufficiently corrupt documents the `getDocument` call will succeed, but fetching even the very first page fails. Currently we only print error messages (in the console) from the `{BaseViewer, PDFThumbnailViewer}.setDocument` methods, but don't actually provide these errors to allow the viewer to handle them properly.
In practice this means that the GENERIC viewer won't display the `errorWrapper`, and in the MOZCENTRAL viewer the *browser* loading indicator is never hidden (since we never unblock the "load" event).
Trying to shadow a non-existent property is always an implementation mistake, since it leads to the `shadow`-call not having any effect.
In PR 14152 I overlooked the fact that it's fairly easy to enforce this during development/testing, since that can help catch e.g. simple spelling bugs.
Currently the `Catalog.metadata` getter only handles errors during parsing, however in a *corrupt* PDF document fetching of the raw /Metadata can obviously fail as well.
Without this patch the `PDFDocumentProxy.getMetadata` method, in the API, can thus fail which it *never* should and this will cause the viewer to not initialize all state as expected.
Fixes one of the documents in issue 14305.
Given that [bug 1336572](https://bugzilla.mozilla.org/show_bug.cgi?id=1336572) was just closed as fixed, thus fixing issue 8019 in Firefox[1], let's add a test-case to enable us to catch any future regressions either in PDF.js or in browsers themselves.
---
[1] It also seems to be working in Google Chrome, although I'm having a slightly difficult time deciphering *exactly* what configurations were affected when looking through issue 8019.
*This patch improves handling of a couple of PDF documents from issue 14303.*
- Update `XRef.indexObjects` to actually clear *all* XRef-caches. Invalid XRef tables *usually* cause issues early enough during parsing that we've not populated the XRef-cache, however to prevent any issues we obviously need to clear that one as well.
- Improve the /Root dictionary validation in `XRef.parse` (PR 9827 follow-up). In addition to checking that a /Pages entry exists, we'll now also check that it can be successfully fetched *and* that it's of the correct type. There's really no point trying to use a /Root dictionary that e.g. `Catalog.toplevelPagesDict` will reject, and this way we'll be able to fallback to indexing the objects in corrupt documents.
- Throw an `InvalidPDFException`, rather than a general `FormatError`, in `XRef.parse` when no usable /Root dictionary could be found. That really seems more appropriate overall, since all attempts at parsing/recovery have failed. (This part of the patch is API-observable, hence the tag.)
With these changes, two existing test-cases are improved and the unit-tests are updated/re-factored to highlight that. In particular `GHOSTSCRIPT-698804-1-fuzzed.pdf` will now both load and "render" correctly, whereas `poppler-395-0-fuzzed.pdf` will now fail immediately upon loading (rather than *appearing* to work).
*Please note:* This is similar to the method that existed prior to PR 3848, but the new method will *only* be used as a fallback when parsing of corrupt PDF documents.
The implementation in PR 14311 unfortunately turned out to be *way* too simplistic, as evident by the recently added test-files in issue 14303, since it may *cause* infinite loops in `PDFDocument.checkLastPage` for some corrupt PDF documents.[1]
To avoid this, the easiest solution that I could come up with was to fallback to eagerly parsing the *entire* /Pages-tree when the /Count-entry validation fails during document initialization.
Fixes *at least* two of the issues listed in issue 14303, namely the `poppler-395-0.pdf...` and `GHOSTSCRIPT-698804-1.pdf...` documents.
---
[1] The whole point of PR 14311 was obviously to *get rid of* infinte loops during document initialization, not to introduce any more of those.
Given that we're able to "render" this document, let's extend the unit-test to actually check that we're able to obtain the operatorList; although given the overall issues in the document it'll be empty.
This was added in PR 14311, but given that I completely missed to update the `PDFDocument.getPage` signature accordingly it's completely unused.
Given that things work just as fine as-is, let's simply remove that optional parameter for now; sorry about the churn here!
This only applies to severely corrupt documents, where it's possible that the `Parser` throws when we try to access e.g. a /Kids-entry in the /Pages-tree.
Fixes two of the issues listed in issue 14303, namely the `poppler-742-0.pdf...` and `poppler-937-0.pdf...` documents.
Given that Node.js doesn't support Workers, general PDF.js performance will be worse when compared to browsers. In an attempt to improve at least memory usage a little bit, update the Node.js examples to release page resources once parsing is done for that page.
This patch is essentially a continuation of PR 11263, which tried to improve loading/initialization performance of *very* large/long documents.
Note that browsers, in general, don't handle a huge amount of DOM-elements very well, with really poor (e.g. sluggish scrolling) performance once the number gets "large". Furthermore, at least in Firefox, it seems that DOM-elements towards the bottom of a HTML-page can effectively be ignored; for the PDF.js viewer that means that pages at the end of the document can become impossible to access.
Hence, in order to improve things for these *very* large/long documents, this patch will now enforce usage of the (recently added) PAGE-scrolling mode for these documents. As implemented, this will only happen once the number of pages *exceed* 15000 (which is hopefully rare in practice).
While this might feel a bit jarring to users being *forced* to use PAGE-scrolling, it seems all things considered like a better idea to ensure that the entire document actually remains accessible and with (hopefully) more acceptable performance.
Fixes [bug 1588435](https://bugzilla.mozilla.org/show_bug.cgi?id=1588435), to the extent that doing so is possible since the document contains 25560 pages (and is 197 MB large).
This was added on the assumption that the viewer would (eventually) start using the `PDFSinglePageViewer` for e.g. PAGE-scrolling mode and PresentationMode. However, having both a `PDFViewer` and a `PDFSinglePageViewer` side-by-side in the viewer would've been tricky to implement well, which is why PR 14112 implemented PAGE-scrolling for the general `BaseViewer` instead.
Given that the default viewer is no longer (potentially) going to use `PDFSinglePageViewer`, there's code in the `SecondaryToolbar` (and related CSS rules) which is now unnecessary.
Since NPM 7, which is over a year old now since it released in October
2020, NPM automatically transforms lock files from version 1 to version
2. In the NPM 7 release notes they reported:
"One change to take note of is the new lockfile format, which is
backwards compatible with npm 6 users. The lockfile v2 unlocks the
ability to do deterministic and reproducible builds to produce a
package tree."
Not only is this change backwards compatible (so older versions of NPM
will still be able to install everything as expected), reproducability
is also a nice property to have and modern NPM versions will otherwise
constantly do the conversion anyway, causing contributors to explicitly
have to revert the change. Therefore, I believe we should do this now
since it doesn't break backwards compatibility for consumers of this
file. It only means that producers of this file (i.e., us contributors)
need to use at least NPM 7 or higher (as of writing NPM 8 is even
available). According to https://nodejs.org/en/download/releases/ this
means contributors should at least run Node.js 15.0.0, while 17.1.0 is
the most recent as of writing, so to me that sounds reasonable to ask.
In Puppeteer 11 we noticed that Firefox doesn't shut down once the tests
are done anymore. I tracked this down to the `page.goto` call, in
`startBrowser`, never resolving anymore. I can only assume that
something changed in Puppeteer, possibly in combination with recent
Firefox Nightly versions, that caused this, but haven't been able to
fully track it down.
However, I did find that the problem is that the `load` event no longer
triggers, so fortunately we can fix the problem by explicitly waiting
for the `domcontentloaded` event instead. In general this change might
even be better since we now wait until the test framework is fully
loaded before we continue. Note that this also still works for the
current Puppeteer version.
I did find two upstream references that appear to track this issue, both
on the Puppeteer side and on the Firefox side, making me further suspect
that the issue is partly on both sides:
- https://github.com/puppeteer/puppeteer/issues/5806
- https://bugzilla.mozilla.org/show_bug.cgi?id=1706353
If a browser cannot be started, we currently get the following log:
`Error while starting firefox: [object Object]`. This is simply an
oversight from the initial Puppeteer integration work since we never got
into this code path before. With this fix the error log becomes more
useful: `Error while starting firefox: connect ECONNREFUSED ::1:45387`
*Please note:* While this patch on its own is sufficient to prevent the worker-thread from hanging, however in combination with PR 14311 these PDF documents will both load *and* render correctly.
Rather than focusing on the particular structure of these PDF documents, it seemed (at least to me) to make sense to try and prevent all circular references when fetching/looking-up data using the XRef table.
To avoid a solution that required tracking the references manually everywhere, the implementation settled on here instead handles that internally in the `XRef.fetch`-method. This should work, since that method *and* the `Parser`/`Lexer`-implementations are completely synchronous.
Note also that the existing `XRef`-caching, used for all data-types *except* Streams, should hopefully help to lessen the performance impact of these changes.
One *potential* problem with these changes could be certain *browser* exceptions, since those are generally not catchable in JavaScript code, however those would most likely "stop" worker-thread parsing anyway (at least I hope so).
Finally, note that I settled on returning dummy-data rather than throwing an exception. This was done to allow parsing, for the rest of the document, to continue such that *one* bad reference doesn't prevent an entire document from loading.
Fixes two of the issues listed in issue 14303, namely the `poppler-91414-0.zip-2.gz-53.pdf` and `poppler-91414-0.zip-2.gz-54.pdf` documents.
*This patch basically extends the approach from PR 10392, by also checking the last page.*
Currently, in e.g. the `Catalog.numPages`-getter, we're simply assuming that if the /Pages-tree has an *integer* /Count entry it must also be correct/valid.
As can be seen in the referenced PDF documents, that entry may be completely bogus which causes general parsing to breaking down elsewhere in the worker-thread (and hanging the browser).
Rather than hoping that the /Count entry is correct, similar to all other data found in PDF documents, we obviously need to validate it. This turns out to be a little less straightforward than one would like, since the only way to do this (as far as I know) is to parse the *entire* /Pages-tree and essentially counting the pages.
To avoid doing that for all documents, this patch tries to take a short-cut by checking if the last page (based on the /Count entry) can be successfully fetched. If so, we assume that the /Count entry is correct and use it as-is, otherwise we'll iterate through (potentially) the *entire* /Pages-tree to determine the number of pages.
Unfortunately these changes will have a number of *somewhat* negative side-effects, please see a possibly incomplete list below, however I cannot see a better way to address this bug.
- This will slow down initial loading/rendering of all documents, at least by some amount, since we now need to fetch/parse more of the /Pages-tree in order to be able to access the *last* page of the PDF documents.
- For poorly generated PDF documents, where the entire /Pages-tree only has *one* level, we'll unfortunately need to fetch/parse the *entire* /Pages-tree to get to the last page. While there's a cache to help reduce repeated data lookups, this will affect initial loading/rendering of *some* long PDF documents,
- This will affect the `disableAutoFetch = true` mode negatively, since we now need to fetch/parse more data during document initialization. While the `disableAutoFetch = true` mode should still be helpful in larger/longer PDF documents, for smaller ones the effect/usefulness may unfortunately be lost.
As one *small* additional bonus, we should now also be able to support opening PDF documents where the /Pages-tree /Count entry is completely invalid (e.g. contains a non-integer value).
Fixes two of the issues listed in issue 14303, namely the `poppler-67295-0.pdf` and `poppler-85140-0.pdf` documents.
Given that not all pages necessarily are being accessed, or that the pages may be accessed out of order, using a `Map` seems like a more appropriate data-structure here.
Furthermore, this patch also adds (currently missing) caching for XFA-documents. Loading a couple of such documents in the viewer, with logging added, shows that we're currently re-creating `Page`-instances unnecessarily for XFA-documents.
