Closed mgyugcha closed 4 years ago
Totals | |
---|---|
Change from base Build 275: | 0.0% |
Covered Lines: | 285 |
Relevant Lines: | 285 |
Thanks, with this 'extraction request' this isuue is solved. We have been working hard with @mgyugcha to fix it, and everything has been satisfactory. Thanks @mgyugcha and congratulations. It is a great contribution.
Great! Not sure if this is the best solution. Best would be to extract the dictionary contents and then either convert to a js Map or search only within the contents. Either way.
By reading the code I don't understand what this commit fixes. If I read through the issue, I do. If you write a test that covers that, it will be obvious when reading the code and code coverage will not be lowered, hence @coveralls won't complain. Please take the time to write a test so this can be merged.
Hi @vbuch, please check the test I did.
Without this update the new trailer contains two startxref
when generating the placeholder, as shown in the picture:
In some PDFs the
/Root
is the last key of the trailer so the search should be with/
and also with>
.Explanation
If this is ignored the rootSlice has unwanted values:
Expected:
Obtained:
This causes the PDF to have an additional startxref: