Next ==== *** ABI changes have been made. build.mk has been updated. * 64-bit windows build, remaining steps - new external-libs have been built and copied into ~/Q/storage/releases/qpdf/external-libs. Release is done in git. Just need to upload when ready. Remember to document that this version is needed for > 2.3.1. - update README-windows.txt docs to indicate that MSVC 2010 is the supported version and to update the information about mingw. * Testing for files > 4GB - Create a PDF from scratch. Each page has a page number as text and an image. The image can be 5000x5000 pixels using 8-bit gray scale. It will be divided into 10 stripes of 500 pixels each. The left and right 500 pixels of each stripe will alternate black and white. The remaining part of the image will have white stripes indicating 1 and black stripes indicating 0 with the most-significant bit on top to indicate the page number. In this way, every page will be unique and will consume approximately 25 megabytes. Creating 200 pages like this will make a file that is 5 GB. - The file will have to have object streams since a regular xref table won't be able to support offsets that large. - A separate test program can create this file and do various manipulations on it. This can be enabled with an environment variable controlled by configure in much the same way image comparison tests are enabled now. The argument to --enable-large-file-test should be a path that has enough disk space to do the tests, probably enough space for two coipes of the file. The test program should also have an interactive mode so we can generate the large file and then look at it with a PDF viewer like Adobe Reader. * Consider adding an example that uses the page APIs, or update the documentation to refer the user to the test suite. Soon ==== * Provide an option to copy encryption parameters from another file. This would make it possible to decrypt a file, manually work with it, and then re-encrypt it using the original encryption parameters including a possibly unknown owner password. * See if I can support the new encryption formats mentioned in the open bug on sourceforge. Check other sourceforge bugs. * Splitting/merging concepts newPDF() could create a PDF with just a trailer, no pages, and a minimal info. Then the page routines could be used to add pages to it. Starting with any pdf, you should be able to copy objects from another pdf. The copy should be smart about never traversing into a /Page or /Pages. We could provide a method of copying objects from one PDF into another. This would do whatever optimization is necessary (maybe just optimizePagesTree) and then traverse the set of objects specified to find all objects referenced by the set. Each of those would be copied over with a table mapping old ID to new ID. This would be done from bottom up most likely disallowing cycles or handling them sanely. Command line could be something like --pages [ --new ] { file [password] numeric-range ... } ... -- The first file referenced would be the one whose other data would be preserved (like trailer, info, encryption, outlines, etc.). --new as first file would just use an empty file as the starting point. Example: to grab pages 1-5 from file1 and 11-15 from file2 --pages file1.pdf 1-5 file2.pdf 11-15 -- To implement this, we would remove all pages from file1 except pages 1 through 5. Then we would take pages 11 through 15 from file2 and add them to a set for transfer. This would end up generating a list of indirect objects. We would copy those objects shallowly to the new PDF keeping track of the mapping and replacing any indirect object keys as appropriate, much like QPDFWriter does. When all the objects are registered, we would add those pages to the result. This approach could work for both splitting and merging. It's possible it could be implemented now without any new APIs, but most of the work should be doable by the library with only a small set of additions. newPDF() QPDFObjectCopier c(qpdf1, qpdf2) QPDFObjectHandle obj = c.copyObject() Without traversing pages, copies all indirect objects referenced by preserving referential integrity and returns an object handle in qpdf2 of the same object. If called multiple times on the same object, retraverses in case there were changes. QPDFObjectHandle obj = c.getMapping() find the object in qpdf2 corresponding to the object from qpdf1. Return the null object if none. General ======= * Look for %PDF header somewhere within the first 1024 bytes of the file. Also accept headers of the form "%!PS−Adobe−N.n PDF−M.m". See Implementation notes 13 and 14 in appendix H of the PDF 1.7 specification. This is bug 3267974. * Update qpdf docs about non-ascii passwords. See thread from 2010-12-07,08 for details. * Look at page splitting. Subramanyam provided a test file; see ../misc/article-threads.pdf. Email Q-Count: 431864 from 2009-11-03. See also "Splitting by Pages" below. * Consider writing a PDF merge utility. With 2.2, it would be possible to have a StreamDataProvider that would allow stream data to be directly copied from one PDF file to another. One possible strategy would be to have a program that adds all the pages of one file to the end of another file. The basic strategy would be to create a table that adds new streams to the original file, mapping the new streams' obj/gen to a stream in the file whose pages are being appended. The StreamDataProvider, when asked, could simply pipe the streams of the file being appended to the provided pipeline and could copy the filter and decode parameters from the original file. Being able to do this requires a lot of the same logic as being able to do splitting, so a general split/merge program would be a great addition. * See whether it's possible to remove the call to flattenScalarReferences. I can't easily figure out why I do it, but removing it causes strange test failures in linearization. I would have to study the optimization and linearization code to figure out why I added this to begin with and what in the code assumes it's the case. For enqueueObject and unparseChild in QPDFWriter, simply removing the checks for indirect scalars seems sufficient. Looking back at the branch in the apex epub repository, before flattening scalar references, there was special case code in QPDFWriter to avoid writing out indirect nulls. It's still not obvious to me why I did it though. To pursue this, remove the call to flattenScalarReferences in QPDFWriter.cc and disable the logic_error exceptions for indirect scalars. Just search for flattenScalarReferences in QPDFWriter.cc since the logic errors have comments that mention flattenScalarReferences. Then run the test suite. Several files that explicitly test flattening of scalar references fail, but the indirect scalars are properly preserved and written. But then there are some linearized files that have a bunch of unreferenced objects that contain scalars. Need to figure out what these are and why they're there. Maybe they're objects that used to be stream lengths. Probably we just need to make sure don't traverse through a stream's /Length stream when enqueueing stream dictionaries. This could potentially happen with any object that QPDFWriter replaces when writing out files. Such objects would be orphaned in the newly written file. This could be fixed, but it may not be worth fixing. If flattenScalarReferences is removed, a new method will be needed for checking PDF files. * See if we can avoid preserving unreferenced objects in object streams even when preserving the object streams. * For debugging linearization bugs, consider adding an option to save pass 1 of linearization. This code is sufficient. Change the interface to allow specification of a pass1 file, which would change the behavior as in this patch. ------------------------------ Index: QPDFWriter.cc =================================================================== --- QPDFWriter.cc (revision 932) +++ QPDFWriter.cc (working copy) @@ -1965,11 +1965,15 @@ // Write file in two passes. Part numbers refer to PDF spec 1.4. + FILE* XXX = 0; for (int pass = 1; pass <= 2; ++pass) { if (pass == 1) { - pushDiscardFilter(); +// pushDiscardFilter(); + XXX = fopen("/tmp/pass1.pdf", "w"); + pushPipeline(new Pl_StdioFile("pass1", XXX)); + activatePipelineStack(); } // Part 1: header @@ -2204,6 +2208,8 @@ // Restore hint offset this->xref[hint_id] = QPDFXRefEntry(1, hint_offset, 0); + fclose(XXX); + XXX = 0; } } } ------------------------------ * Handle embedded files. PDF Reference 1.7 section 3.10, "File Specifications", discusses this. Once we can definitely recongize all embedded files in a docucment, we can update the encryption code to handle it properly. In QPDF_encryption.cc, search for cf_file. Remove exception thrown if cf_file is different from cf_stream, and write code in the stream decryption section to use cf_file instead of cf_stream. In general, add interfaces to get the list of embedded files and to extract them. To handle general embedded files associated with the whole document, follow root -> /Names -> /EmbeddedFiles -> /Names to get to the file specification dictionaries. Then, in each file specification dictionary, follow /EF -> /F to the actual stream. There may be other places file specification dictionaries may appear, and there are also /RF keys with related files, so reread section 3.10 carefully. * The description of Crypt filters is unclear with respect to how to use them to override /StmF for specific streams. I'm not sure whether qpdf will do the right thing for any specific individual streams that might have crypt filters. The specification seems to imply that only embedded file streams and metadata streams can have crypt filters, and there are already special cases in the code to handle those. Most likely, it won't be a problem, but someday someone may find a file that qpdf doesn't work on because of crypt filters. There is an example in the spec of using a crypt filter on a metadata stream. For now, we notice /Crypt filters and decode parameters consistent with the example in the PDF specification, and the right thing happens for metadata filters that happen to be uncompressed or otherwise compressed in a way we can filter. This should handle all normal cases, but it's more or less just a guess since I don't have any test files that actually use stream-specific crypt filters in them. * The second xref stream for linearized files has to be padded only because we need file_size as computed in pass 1 to be accurate. If we were not allowing writing to a pipe, we could seek back to the beginning and fill in the value of /L in the linearization dictionary as an optimization to alleviate the need for this padding. Doing so would require us to pad the /L value individually and also to save the file descriptor and determine whether it's seekable. This is probably not worth bothering with. * The whole xref handling code in the QPDF object allows the same object with more than one generation to coexist, but a lot of logic assumes this isn't the case. Anything that creates mappings only with the object number and not the generation is this way, including most of the interaction between QPDFWriter and QPDF. If we wanted to allow the same object with more than one generation to coexist, which I'm not sure is allowed, we could fix this by changing xref_table. Alternatively, we could detect and disallow that case. In fact, it appears that Adobe reader and other PDF viewing software silently ignores objects of this type, so this is probably not a big deal. * Pl_PNGFilter is only partially implemented. If we ever decoded images, we'd have to finish implementing it along with the other filter decode parameters and types. For just handling xref streams, there's really no need as it wouldn't make sense to use any kind of predictor other than 12 (PNG UP filter). * If we ever want to have check mode check the integrity of the free list, this can be done by looking at the code from prior to the object stream support of 4/5/2008. It's in an if (0) block and there's a comment about it. There's also something about it in qpdf.test -- search for "free table". On the other hand, the value of doing this seems very low since no viewer seems to care, so it's probably not worth it. * QPDFObjectHandle::getPageImages() doesn't notice images in inherited resource dictionaries. See comments in that function. * Based on an idea suggested by user "Atom Smasher", consider providing some mechanism to recover earlier versions of a file embedded prior to appended sections. * From a suggestion in bug 3152169, consisder having an option to re-encode inline images with an ASCII encoding. Splitting by Pages ================== Although qpdf does not currently support splitting a file into pages, the work done for linearization covers almost all the work. To do page splitting. If this functionality is needed, study obj_user_to_objects and object_to_obj_users created in QPDF_optimization for ideas. It's quite possible that the information computed by calculateLinearizationData is actually sufficient to do page splitting in many circumstances. That code knows which objects are used by which pages, though it doesn't do anything page-specific with outlines, thumbnails, page labels, or anything else. Another approach would be to traverse only pages that are being output taking care not to traverse into the pages tree, and then to fabricate a new pages tree. Either way, care must be taken to handle other things such as outlines, page labels, thumbnails, threads, zones, etc. in a sensible way. This may include simply omitting information other than page content.