Script 'mail_helper' called by obssrc
Hello community,
here is the log from the commit of package python-pikepdf for openSUSE:Factory
checked in at 2023-12-28 23:02:03
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/python-pikepdf (Old)
and /work/SRC/openSUSE:Factory/.python-pikepdf.new.28375 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "python-pikepdf"
Thu Dec 28 23:02:03 2023 rev:21 rq:1135318 version:8.10.1
Changes:
--------
--- /work/SRC/openSUSE:Factory/python-pikepdf/python-pikepdf.changes
2023-12-14 22:03:30.796873948 +0100
+++ /work/SRC/openSUSE:Factory/.python-pikepdf.new.28375/python-pikepdf.changes
2023-12-28 23:03:47.335541986 +0100
@@ -1,0 +2,11 @@
+Wed Dec 27 14:00:51 UTC 2023 - Dirk Müller <dmuel...@suse.com>
+
+- update to 8.10.1:
+ * Rebuilt with QPDF 11.6.4.
+ * Replaced use of a custom C++ logger with sharing QPDF's.
+ It is still relayed to the Python logger.
+ * Added a simpler API for adding attachments from bytes data.
+ * Deprecated use of Object.parse(str) in favor of
+ Object.parse(bytes).
+
+-------------------------------------------------------------------
Old:
----
pikepdf-8.9.0.tar.gz
New:
----
pikepdf-8.10.1.tar.gz
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Other differences:
------------------
++++++ python-pikepdf.spec ++++++
--- /var/tmp/diff_new_pack.F5kIK0/_old 2023-12-28 23:03:48.007566547 +0100
+++ /var/tmp/diff_new_pack.F5kIK0/_new 2023-12-28 23:03:48.007566547 +0100
@@ -19,7 +19,7 @@
%{?sle15_python_module_pythons}
Name: python-pikepdf
-Version: 8.9.0
+Version: 8.10.1
Release: 0
Summary: Read and write PDFs with Python, powered by qpdf
License: MPL-2.0
++++++ pikepdf-8.9.0.tar.gz -> pikepdf-8.10.1.tar.gz ++++++
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/PKG-INFO new/pikepdf-8.10.1/PKG-INFO
--- old/pikepdf-8.9.0/PKG-INFO 2023-12-11 00:26:49.663727000 +0100
+++ new/pikepdf-8.10.1/PKG-INFO 2023-12-17 10:33:05.458814400 +0100
@@ -1,6 +1,6 @@
Metadata-Version: 2.1
Name: pikepdf
-Version: 8.9.0
+Version: 8.10.1
Summary: Read and write PDFs with Python, powered by qpdf
Author-email: "James R. Barlow" <ja...@purplerock.ca>
License: MPL-2.0
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/docs/conf.py
new/pikepdf-8.10.1/docs/conf.py
--- old/pikepdf-8.9.0/docs/conf.py 2023-12-11 00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/docs/conf.py 2023-12-17 10:31:00.000000000 +0100
@@ -92,7 +92,7 @@
# |version| and |release|, also used in various other places throughout the
# built documents.
-release = "8.9.0"
+release = "8.10.1"
version = '.'.join(release.split('.')[:2])
# The language for content autogenerated by Sphinx. Refer to documentation
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/docs/releasenotes/version8.rst
new/pikepdf-8.10.1/docs/releasenotes/version8.rst
--- old/pikepdf-8.9.0/docs/releasenotes/version8.rst 2023-12-11
00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/docs/releasenotes/version8.rst 2023-12-17
10:31:00.000000000 +0100
@@ -1,3 +1,25 @@
+v8.10.1
+=======
+
+- Rebuilt with QPDF 11.6.4.
+- Replaced use of a custom C++ logger with sharing QPDF's. It is still relayed
to
+ the Python logger.
+- Added a simpler API for adding attachments from bytes data.
+- Deprecated use of Object.parse(str) in favor of Object.parse(bytes).
+
+v8.10.0
+=======
+
+- Fixed a performance regression when appending thousands of pages from one
PDF to
+ another.
+- Fixed some obscure issues with iterators over ``Pdf.pages`` that would have
led
+ to incorrect or unintuitive behavior, like partial iteration not being
accounted
+ for.
+- Using the ``Pdf.pages`` API to insert objects other ``pikepdf.Pdf`` is now
+ deprecated. Previously, we accepted ``pikepdf.Dictionary`` that had its
``/Type``
+ set to ``/Page``. Now, one must wrap these dictionaries in
``pikepdf.Page()``.
+- Added type hints that ``pikepdf.Object`` can be implicitly converted to float
+ and int.
v8.9.0
======
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/pyproject.toml
new/pikepdf-8.10.1/pyproject.toml
--- old/pikepdf-8.9.0/pyproject.toml 2023-12-11 00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/pyproject.toml 2023-12-17 10:31:00.000000000 +0100
@@ -7,7 +7,7 @@
[project]
name = "pikepdf"
-version = "8.9.0"
+version = "8.10.1"
description = "Read and write PDFs with Python, powered by qpdf"
readme = "README.md"
requires-python = ">=3.8"
@@ -114,7 +114,7 @@
[tool.cibuildwheel.environment]
QPDF_MIN_VERSION = "11.5.0"
-QPDF_VERSION = "11.6.3"
+QPDF_VERSION = "11.6.4"
QPDF_PATTERN =
"https://github.com/qpdf/qpdf/releases/download/vVERSION/qpdf-VERSION.tar.gz";
[tool.cibuildwheel.linux]
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/embeddedfiles.cpp
new/pikepdf-8.10.1/src/core/embeddedfiles.cpp
--- old/pikepdf-8.9.0/src/core/embeddedfiles.cpp 2023-12-11
00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/src/core/embeddedfiles.cpp 2023-12-17
10:31:00.000000000 +0100
@@ -15,6 +15,33 @@
#include "pikepdf.h"
#include "pipeline.h"
+QPDFFileSpecObjectHelper create_filespec(QPDF &q,
+ py::bytes data,
+ std::string description,
+ std::string filename,
+ std::string mime_type,
+ std::string creation_date,
+ std::string mod_date,
+ QPDFObjectHandle relationship)
+{
+ auto efstream = QPDFEFStreamObjectHelper::createEFStream(q,
std::string(data));
+ auto filespec = QPDFFileSpecObjectHelper::createFileSpec(q, filename,
efstream);
+
+ if (!description.empty())
+ filespec.setDescription(description);
+ if (!mime_type.empty())
+ efstream.setSubtype(mime_type);
+ if (!creation_date.empty())
+ efstream.setCreationDate(creation_date);
+ if (!mod_date.empty())
+ efstream.setModDate(mod_date);
+
+ if (relationship.isName()) {
+ filespec.getObjectHandle().replaceKey("/AFRelationship", relationship);
+ }
+ return filespec;
+}
+
void init_embeddedfiles(py::module_ &m)
{
py::class_<QPDFFileSpecObjectHelper,
@@ -28,24 +55,14 @@
std::string creation_date,
std::string mod_date,
QPDFObjectHandle &relationship) {
- auto efstream =
- QPDFEFStreamObjectHelper::createEFStream(q, std::string(data));
- auto filespec =
- QPDFFileSpecObjectHelper::createFileSpec(q, filename,
efstream);
-
- if (!description.empty())
- filespec.setDescription(description);
- if (!mime_type.empty())
- efstream.setSubtype(mime_type);
- if (!creation_date.empty())
- efstream.setCreationDate(creation_date);
- if (!mod_date.empty())
- efstream.setModDate(mod_date);
-
- if (relationship.isName()) {
- filespec.getObjectHandle().replaceKey("/AFRelationship",
relationship);
- }
- return filespec;
+ return create_filespec(q,
+ data,
+ description,
+ filename,
+ mime_type,
+ creation_date,
+ mod_date,
+ relationship);
}),
py::keep_alive<0, 1>(), // LCOV_EXCL_LINE
py::arg("q"),
@@ -119,6 +136,18 @@
py::class_<QPDFEmbeddedFileDocumentHelper>(m, "Attachments")
.def_property_readonly(
"_has_embedded_files",
&QPDFEmbeddedFileDocumentHelper::hasEmbeddedFiles)
+ .def("_attach_data",
+ [](QPDFEmbeddedFileDocumentHelper &efdh, py::str key, py::bytes
data) {
+ auto ef = create_filespec(efdh.getQPDF(),
+ std::string(data),
+ std::string(""),
+ std::string(key),
+ std::string(""),
+ std::string(""),
+ std::string(""),
+ QPDFObjectHandle::newName("/Unspecified"));
+ efdh.replaceEmbeddedFile(key, ef);
+ })
.def("_get_all_filespecs",
&QPDFEmbeddedFileDocumentHelper::getEmbeddedFiles,
py::return_value_policy::reference_internal)
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/logger.cpp
new/pikepdf-8.10.1/src/core/logger.cpp
--- old/pikepdf-8.9.0/src/core/logger.cpp 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/core/logger.cpp 2023-12-17 10:31:00.000000000
+0100
@@ -4,8 +4,6 @@
#include "pikepdf.h"
#include <qpdf/QPDFLogger.hh>
-static auto pikepdf_logger = QPDFLogger::create();
-
// Pipeline to relay QPDF log messages to Python logging module
// This is a sink - cannot pass to other pipeline objects
class Pl_PythonLogger : public Pipeline {
@@ -47,7 +45,7 @@
std::shared_ptr<QPDFLogger> get_pikepdf_logger()
{
// All QPDFs can use the same logger
- return pikepdf_logger;
+ return QPDFLogger::defaultLogger();
}
void init_logger(py::module_ &m)
@@ -61,6 +59,7 @@
std::shared_ptr<Pipeline> pl_log_error = std::make_shared<Pl_PythonLogger>(
"QPDF to Python logging pipeline", py_logger, "error");
+ auto pikepdf_logger = get_pikepdf_logger();
pikepdf_logger->setInfo(pl_log_info);
pikepdf_logger->setWarn(pl_log_warn);
pikepdf_logger->setError(pl_log_error);
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/object.cpp
new/pikepdf-8.10.1/src/core/object.cpp
--- old/pikepdf-8.9.0/src/core/object.cpp 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/core/object.cpp 2023-12-17 10:31:00.000000000
+0100
@@ -758,7 +758,17 @@
.def_property_readonly("objgen", &object_get_objgen)
.def_static(
"parse",
- [](std::string const &stream, std::string const &description) {
+ [](py::bytes stream, py::str description) {
+ return QPDFObjectHandle::parse(
+ std::string(stream), std::string(description));
+ },
+ py::arg("stream"),
+ py::arg("description") = "")
+ .def_static(
+ "parse",
+ [](py::str stream, std::string const &description) {
+ python_warning("pikepdf.Object.parse(str) is deprecated; use
bytes.",
+ PyExc_DeprecationWarning);
return QPDFObjectHandle::parse(stream, description);
},
py::arg("stream"),
@@ -863,7 +873,7 @@
.def(
"__eq__",
[](QPDFObjectHelper &self, QPDFObjectHelper &other) {
- // Pages that are copies
+ // Object helpers are equal if their object handles are equal
return objecthandle_equal(
self.getObjectHandle(), other.getObjectHandle());
},
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/object_convert.cpp
new/pikepdf-8.10.1/src/core/object_convert.cpp
--- old/pikepdf-8.9.0/src/core/object_convert.cpp 2023-12-11
00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/src/core/object_convert.cpp 2023-12-17
10:31:00.000000000 +0100
@@ -43,13 +43,9 @@
{
StackGuard sg(" array_builder");
std::vector<QPDFObjectHandle> result;
- int narg = 0;
for (const auto &item : iter) {
- narg++;
-
- auto value = objecthandle_encode(item);
- result.push_back(value);
+ result.emplace_back(objecthandle_encode(item));
}
return result;
}
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/parsers.cpp
new/pikepdf-8.10.1/src/core/parsers.cpp
--- old/pikepdf-8.9.0/src/core/parsers.cpp 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/core/parsers.cpp 2023-12-17 10:31:00.000000000
+0100
@@ -58,8 +58,6 @@
}
break;
}
- case qpdf_object_type_e::ot_stream:
- throw py::type_error("Streams are not allowed in content stream
instructions");
default: {
throw py::type_error("Only scalar types, arrays, and dictionaries are
allowed "
"in content streams.");
@@ -271,13 +269,11 @@
{
py::class_<ContentStreamInstruction>(m, "ContentStreamInstruction")
.def(py::init<const ContentStreamInstruction &>())
- .def(py::init([](ObjectList operands, QPDFObjectHandle operator_) {
- return ContentStreamInstruction(operands, operator_);
- }))
+ .def(py::init<ObjectList, QPDFObjectHandle>())
.def(py::init([](py::iterable operands, QPDFObjectHandle operator_) {
ObjectList newlist;
for (auto &item : operands) {
- newlist.push_back(objecthandle_encode(item));
+ newlist.emplace_back(objecthandle_encode(item));
}
return ContentStreamInstruction(newlist, operator_);
}))
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/pikepdf.cpp
new/pikepdf-8.10.1/src/core/pikepdf.cpp
--- old/pikepdf-8.9.0/src/core/pikepdf.cpp 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/core/pikepdf.cpp 2023-12-17 10:31:00.000000000
+0100
@@ -136,6 +136,17 @@
init_rectangle(m);
init_tokenfilter(m);
+ auto m_test = m.def_submodule("_test", "pikepdf._core test functions");
+ m_test
+ .def(
+ "fopen_nonexistent_file",
+ []() -> void { (void)QUtil::safe_fopen("does_not_exist__42",
"rb"); },
+ "Used to test that C++ system error -> Python exception
propagation works.")
+ .def(
+ "log_info",
+ [](std::string s) { return get_pikepdf_logger()->info(s); },
+ "Used to test routing of QPDF's logger to Python logging.");
+
// -- Module level functions --
m.def("utf8_to_pdf_doc",
[](py::str utf8, char unknown) {
@@ -148,17 +159,9 @@
return py::str(QUtil::pdf_doc_to_utf8(pdfdoc));
})
.def(
- "_test_file_not_found",
- []() -> void { (void)QUtil::safe_fopen("does_not_exist__42",
"rb"); },
- "Used to test that C++ system error -> Python exception
propagation works.")
- .def(
"_translate_qpdf_logic_error",
[](std::string s) { return translate_qpdf_logic_error(s).first; },
"Used to test interpretation of QPDF errors.")
- .def(
- "_log_info",
- [](std::string s) { return get_pikepdf_logger()->info(s); },
- "Used to test routing of QPDF's logger to Python logging.")
.def("set_decimal_precision",
[](uint prec) {
DECIMAL_PRECISION = prec;
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/qpdf_pagelist.cpp
new/pikepdf-8.10.1/src/core/qpdf_pagelist.cpp
--- old/pikepdf-8.9.0/src/core/qpdf_pagelist.cpp 2023-12-11
00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/src/core/qpdf_pagelist.cpp 2023-12-17
10:31:00.000000000 +0100
@@ -44,6 +44,7 @@
if (!slice.compute(this->count(), &start, &stop, &step, &slicelength))
throw py::error_already_set(); // LCOV_EXCL_LINE
std::vector<QPDFPageObjectHelper> result;
+ result.reserve(slicelength);
for (py::size_t i = 0; i < slicelength; ++i) {
auto oh = this->get_page(start);
result.push_back(oh);
@@ -134,8 +135,15 @@
py::size_t PageList::count() { return this->doc.getAllPages().size(); }
-void PageList::try_insert_qpdfobject_as_page(py::size_t index, py::handle obj)
+QPDFPageObjectHelper PageList::page_from_object(py::handle obj)
{
+ try {
+ auto page = obj.cast<QPDFPageObjectHelper>();
+ return page;
+ } catch (py::cast_error &) {
+ // Perhaps obj is a dictionary with Type=Name.Page
+ }
+
QPDFObjectHandle oh, indirect_oh;
try {
oh = obj.cast<QPDFObjectHandle>();
@@ -144,6 +152,10 @@
"nor pikepdf.Dictionary with Type=Name.Page");
}
+ python_warning("Implicit conversion of pikepdf.Dictionary to pikepdf.Page
is "
+ "deprecated. Use pikepdf.Page(dictionary) instead.",
+ PyExc_DeprecationWarning);
+
bool copied = false;
try {
if (!oh.getOwningQPDF()) {
@@ -161,8 +173,7 @@
"to insert this as a page: ") +
objecthandle_repr(oh));
}
- auto page = QPDFPageObjectHelper(indirect_oh);
- this->insert_page(index, page);
+ return QPDFPageObjectHelper(indirect_oh);
} catch (std::runtime_error &) {
// If we created a new temporary indirect object to hold the page, and
// failed to insert, delete the object we created as best we can.
@@ -181,22 +192,33 @@
this->insert_page(index, poh);
return;
} catch (py::cast_error &) {
- this->try_insert_qpdfobject_as_page(index, obj);
+ auto page = this->page_from_object(obj);
+ this->insert_page(index, page);
return;
}
}
void PageList::insert_page(py::size_t index, QPDFPageObjectHelper page)
{
- auto doc = QPDFPageDocumentHelper(*this->qpdf);
if (index != this->count()) {
auto refpage = this->get_page(index);
- doc.addPageAt(page, true, refpage);
+ this->doc.addPageAt(page, true, refpage);
} else {
- doc.addPage(page, false);
+ this->doc.addPage(page, false);
}
}
+void PageList::append_page(py::handle obj)
+{
+ auto page = this->page_from_object(obj);
+ this->doc.addPage(page, false);
+}
+
+void PageList::append_page(QPDFPageObjectHelper page)
+{
+ this->doc.addPage(page, false);
+}
+
QPDFPageObjectHelper from_objgen(QPDF &q, QPDFObjGen og)
{
auto h = q.getObjectByObjGen(og);
@@ -205,8 +227,22 @@
return QPDFPageObjectHelper(h);
}
+QPDFPageObjectHelper PageListIterator::next()
+{
+ if (this->index >= this->pages.size()) {
+ throw py::stop_iteration();
+ }
+ auto page = this->pages.at(this->index);
+ this->index++;
+ return page;
+}
+
void init_pagelist(py::module_ &m)
{
+ py::class_<PageListIterator>(m, "_PageListIterator")
+ .def("__iter__", [](PageListIterator &it) { return it; })
+ .def("__next__", &PageListIterator::next);
+
py::class_<PageList>(m, "PageList")
.def(
"__getitem__",
@@ -238,13 +274,12 @@
return pl.get_page(pnum - 1);
},
py::arg("pnum"))
- .def("__iter__", [](PageList &pl) { return PageList(pl.qpdf, 0); })
- .def("__next__",
+ .def(
+ "__iter__",
[](PageList &pl) {
- if (pl.iterpos < pl.count())
- return pl.get_page(pl.iterpos++);
- throw py::stop_iteration();
- })
+ return PageListIterator{pl, 0};
+ },
+ py::keep_alive<0, 1>())
.def(
"insert",
[](PageList &pl, py::ssize_t index, py::object obj) {
@@ -256,31 +291,24 @@
.def("reverse",
[](PageList &pl) {
py::slice ordinary_indices(0, pl.count(), 1);
- py::int_ step(-1);
- py::slice reversed = py::reinterpret_steal<py::slice>(
- PySlice_New(Py_None, Py_None, step.ptr()));
+ py::slice reversed{{}, {}, -1};
py::list reversed_pages = pl.get_pages(reversed);
pl.set_pages_from_iterable(ordinary_indices, reversed_pages);
})
.def(
"append",
- [](PageList &pl, QPDFPageObjectHelper &page) {
- pl.insert_page(pl.count(), page);
- },
+ [](PageList &pl, QPDFPageObjectHelper &page) {
pl.append_page(page); },
py::arg("page"))
.def(
"append",
- [](PageList &pl, py::handle page) { pl.insert_page(pl.count(),
page); },
+ [](PageList &pl, py::handle page) { pl.append_page(page); },
py::arg("page"))
.def(
"extend",
[](PageList &pl, PageList &other) {
- auto other_count = other.count();
- for (decltype(other_count) i = 0; i < other_count; i++) {
- if (other_count != other.count())
- throw py::value_error(
- "source page list modified during iteration");
- pl.insert_page(pl.count(), other.get_page(i));
+ auto other_pages = other.doc.getAllPages();
+ for (auto &page : other_pages) {
+ pl.append_page(page);
}
},
py::arg("other"))
@@ -289,21 +317,30 @@
[](PageList &pl, py::iterable iterable) {
py::iterator it = iterable.attr("__iter__")();
while (it != py::iterator::sentinel()) {
- // assert_pyobject_is_page_obj(*it);
assert_pyobject_is_page_helper(*it);
- pl.insert_page(pl.count(), *it);
+ pl.append_page(*it);
++it;
}
},
py::arg("iterable"))
.def("remove",
- [](PageList &pl, py::kwargs kwargs) {
- auto pnum = kwargs["p"].cast<py::ssize_t>();
+ [](PageList &pl, QPDFPageObjectHelper &page) {
+ try {
+ pl.doc.removePage(page);
+ } catch (const QPDFExc &e) {
+ throw py::value_error("Page is not referenced in the PDF");
+ }
+ })
+ .def(
+ "remove",
+ [](PageList &pl, py::ssize_t pnum) {
if (pnum <= 0) // Indexing past end is checked in .get_page
throw py::index_error(
"page access out of range in 1-based indexing");
pl.delete_page(pnum - 1);
- })
+ },
+ py::kw_only(),
+ py::arg("p"))
.def("index",
[](PageList &pl, const QPDFObjectHandle &h) {
return page_index(*pl.qpdf, h);
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/core/qpdf_pagelist.h
new/pikepdf-8.10.1/src/core/qpdf_pagelist.h
--- old/pikepdf-8.9.0/src/core/qpdf_pagelist.h 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/core/qpdf_pagelist.h 2023-12-17 10:31:00.000000000
+0100
@@ -14,8 +14,7 @@
class PageList { // LCOV_EXCL_LINE
public:
- PageList(std::shared_ptr<QPDF> q, py::size_t iterpos = 0)
- : iterpos(iterpos), qpdf(q), doc(*qpdf){};
+ PageList(std::shared_ptr<QPDF> q) : qpdf(q), doc(*qpdf){};
QPDFPageObjectHelper get_page(py::size_t index);
py::list get_pages(py::slice slice);
@@ -26,13 +25,26 @@
py::size_t count();
void insert_page(py::size_t index, py::handle obj);
void insert_page(py::size_t index, QPDFPageObjectHelper page);
+ void append_page(py::handle obj);
+ void append_page(QPDFPageObjectHelper page);
public:
- py::size_t iterpos;
std::shared_ptr<QPDF> qpdf;
QPDFPageDocumentHelper doc;
private:
std::vector<QPDFPageObjectHelper> get_page_objs_impl(py::slice slice);
- void try_insert_qpdfobject_as_page(py::size_t index, py::handle obj);
+ QPDFPageObjectHelper page_from_object(py::handle obj);
};
+
+class PageListIterator { // LCOV_EXCL_LINE
+public:
+ PageListIterator(PageList &pl, size_t index)
+ : pl(pl), index(index), pages(pl.doc.getAllPages()){};
+ QPDFPageObjectHelper next();
+
+private:
+ PageList &pl;
+ size_t index;
+ std::vector<QPDFPageObjectHelper> pages;
+};
\ No newline at end of file
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/pikepdf/_core.pyi
new/pikepdf-8.10.1/src/pikepdf/_core.pyi
--- old/pikepdf-8.9.0/src/pikepdf/_core.pyi 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/pikepdf/_core.pyi 2023-12-17 10:31:00.000000000
+0100
@@ -467,9 +467,11 @@
def __delitem__(self, name: str | Name | int) -> None: ...
def __dir__(self) -> list: ...
def __eq__(self, other: Any) -> bool: ...
+ def __float__(self) -> float: ...
def __getattr__(self, name: str) -> Object: ...
def __getitem__(self, name: str | Name | int) -> Object: ...
def __hash__(self) -> int: ...
+ def __int__(self) -> int: ...
def __iter__(self) -> Iterable[Object]: ...
def __len__(self) -> int: ...
def __setattr__(self, name: str, value: Any) -> None: ...
@@ -817,15 +819,36 @@
class Attachments(MutableMapping[str, AttachedFileSpec]):
"""Exposes files attached to a PDF.
+ If a file is attached to a PDF, it is exposed through this interface.
+ For example ``p.attachments['readme.txt']`` would return a
+ :class:`pikepdf._core.AttachedFileSpec` that describes the attached file,
+ if a file were attached under that name.
+ ``p.attachments['readme.txt'].get_file()`` would return a
+ :class:`pikepdf._core.AttachedFile`, an archaic intermediate object to
support
+ different versions of the file for different platforms. Typically one
+ just calls ``p.attachments['readme.txt'].read_bytes()`` to get the
+ contents of the file.
+
This interface provides access to any files that are attached to this PDF,
exposed as a Python :class:`collections.abc.MutableMapping` interface.
The keys (virtual filenames) are always ``str``, and values are always
:class:`pikepdf.AttachedFileSpec`.
+ To create a new attached file, use
+ :meth:`pikepdf._core.AttachedFileSpec.from_filepath`
+ to create a :class:`pikepdf._core.AttachedFileSpec` and then assign it to
the
+ :attr:`pikepdf.Pdf.attachments` mapping. If the file is in memory, use
+ ``p.attachments['test.pdf'] = b'binary data'``.
+
Use this interface through :attr:`pikepdf.Pdf.attachments`.
.. versionadded:: 3.0
+
+ .. versionchanged:: 8.10.1
+ Added convenience interface for directly loading attached files, e.g.
+ ``pdf.attachments['/test.pdf'] = b'binary data'``. Prior to this
release,
+ there was no way to attach data in memory as a file.
"""
def __contains__(self, k: object) -> bool: ...
@@ -834,7 +857,7 @@
def __getitem__(self, k: str) -> AttachedFileSpec: ...
def __iter__(self) -> Iterator[str]: ...
def __len__(self) -> int: ...
- def __setitem__(self, k: str, v: AttachedFileSpec): ...
+ def __setitem__(self, k: str, v: AttachedFileSpec | bytes): ...
def __init__(self, *args, **kwargs) -> None: ...
def _add_replace_filespec(self, arg0: str, arg1: AttachedFileSpec) ->
None: ...
def _get_all_filespecs(self) -> dict[str, AttachedFileSpec]: ...
@@ -1341,10 +1364,11 @@
"""For accessing pages in a PDF.
A ``list``-like object enumerating a range of pages in a
:class:`pikepdf.Pdf`.
- It may be all of the pages or a subset.
+ It may be all of the pages or a subset. Obtain using
:attr:`pikepdf.Pdf.pages`.
+
+ See :class:`pikepdf.Page` for accessing individual pages.
"""
- def __init__(self, *args, **kwargs) -> None: ...
def append(self, page: Page) -> None:
"""Add another page to the end.
@@ -1372,7 +1396,7 @@
Raises an exception if no page matches.
"""
- def index(self, page: Page | Object) -> int:
+ def index(self, page: Page) -> int:
"""Given a page, find the index.
That is, returns ``n`` such that ``pdf.pages[n] == this_page``.
@@ -1398,29 +1422,33 @@
function does not account for that. Use :attr:`pikepdf.Page.label`
to get the page label for a page.
"""
- def remove(self, *, p: int) -> None:
- """Remove a page (using 1-based numbering).
+ def remove(self, page: Page | None = None, *, p: int) -> None:
+ """Remove a page.
Args:
- p: 1-based page number
+ page: If page is not None, remove that page.
+ p: 1-based page number to remove, if page is None.
"""
def reverse(self) -> None:
"""Reverse the order of pages."""
@overload
- def __delitem__(self, arg0: int) -> None: ...
+ def __delitem__(self, idx: int) -> None: ...
@overload
- def __delitem__(self, arg0: slice) -> None: ...
+ def __delitem__(self, sl: slice) -> None: ...
@overload
- def __getitem__(self, arg0: int) -> Page: ...
+ def __getitem__(self, idx: int) -> Page: ...
@overload
- def __getitem__(self, arg0: slice) -> list[Page]: ...
- def __iter__(self) -> PageList: ...
+ def __getitem__(self, sl: slice) -> list[Page]: ...
+ def __iter__(self) -> _PageListIterator: ...
def __len__(self) -> int: ...
- def __next__(self) -> Page: ...
@overload
- def __setitem__(self, arg0: int, arg1: Page) -> None: ...
+ def __setitem__(self, idx: int, page: Page) -> None: ...
@overload
- def __setitem__(self, arg0: slice, arg1: Iterable[Page]) -> None: ...
+ def __setitem__(self, sl: slice, pages: Iterable[Page]) -> None: ...
+
+class _PageListIterator:
+ def __iter__(self) -> _PageListIterator: ...
+ def __next__(self) -> Page: ...
class Pdf:
def _repr_mimebundle_(include: Any = ..., exclude: Any = ...) -> Any:
@@ -2771,7 +2799,6 @@
def _new_string_utf8(s: str) -> String:
"""Low-level function to construct a PDF String object from UTF-8 bytes."""
-def _test_file_not_found(*args, **kwargs) -> Any: ...
def _translate_qpdf_logic_error(arg0: str) -> str: ...
def get_decimal_precision() -> int:
"""Set the number of decimal digits to use when converting floats."""
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/pikepdf/_methods.py
new/pikepdf-8.10.1/src/pikepdf/_methods.py
--- old/pikepdf-8.9.0/src/pikepdf/_methods.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/pikepdf/_methods.py 2023-12-17 10:31:00.000000000
+0100
@@ -55,7 +55,7 @@
Numeric = TypeVar('Numeric', int, float, Decimal)
-def _single_page_pdf(page) -> bytes:
+def _single_page_pdf(page: Page) -> bytes:
"""Construct a single page PDF from the provided page in memory."""
pdf = Pdf.new()
pdf.pages.append(page)
@@ -666,7 +666,7 @@
bundle = {k for k in bundle if k in include}
if exclude:
bundle = {k for k in bundle if k not in exclude}
- pagedata = _single_page_pdf(self.obj)
+ pagedata = _single_page_pdf(self)
if 'application/pdf' in bundle:
data['application/pdf'] = pagedata
if 'image/png' in bundle:
@@ -700,7 +700,9 @@
raise KeyError(k)
return filespec
- def __setitem__(self, k: str, v: AttachedFileSpec) -> None:
+ def __setitem__(self, k: str, v: AttachedFileSpec | bytes) -> None:
+ if isinstance(v, bytes):
+ return self._attach_data(k, v)
if not v.filename:
v.filename = k
return self._add_replace_filespec(k, v)
@@ -715,7 +717,9 @@
yield from self._get_all_filespecs()
def __repr__(self):
- return f"<pikepdf._core.Attachments with {len(self)} attached files>"
+ if len(self) == 0:
+ return "<pikepdf._core.Attachments no attached files>"
+ return f"<pikepdf._core.Attachments: {list(self)}>"
@augments(AttachedFileSpec)
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/pikepdf/_version.py
new/pikepdf-8.10.1/src/pikepdf/_version.py
--- old/pikepdf-8.9.0/src/pikepdf/_version.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/src/pikepdf/_version.py 2023-12-17 10:31:00.000000000
+0100
@@ -3,4 +3,4 @@
from __future__ import annotations
-__version__ = "8.9.0"
+__version__ = "8.10.1"
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/src/pikepdf.egg-info/PKG-INFO
new/pikepdf-8.10.1/src/pikepdf.egg-info/PKG-INFO
--- old/pikepdf-8.9.0/src/pikepdf.egg-info/PKG-INFO 2023-12-11
00:26:49.000000000 +0100
+++ new/pikepdf-8.10.1/src/pikepdf.egg-info/PKG-INFO 2023-12-17
10:33:05.000000000 +0100
@@ -1,6 +1,6 @@
Metadata-Version: 2.1
Name: pikepdf
-Version: 8.9.0
+Version: 8.10.1
Summary: Read and write PDFs with Python, powered by qpdf
Author-email: "James R. Barlow" <ja...@purplerock.ca>
License: MPL-2.0
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_attachments.py
new/pikepdf-8.10.1/tests/test_attachments.py
--- old/pikepdf-8.9.0/tests/test_attachments.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_attachments.py 2023-12-17
10:31:00.000000000 +0100
@@ -33,7 +33,7 @@
assert len(pal.attachments) == 1, "attachment count not incremented"
assert 'rle.pdf' in pal.attachments, "attachment filename not registered"
- assert 'attached' in repr(pal.attachments)
+ assert 'rle.pdf' in repr(pal.attachments), "attachment filename not
enumerated"
pal.save(outpdf)
@@ -157,3 +157,9 @@
assert 'foo' not in repr(fs)
assert 'AttachedFile' in repr(fs.get_file())
assert fs.relationship == Name.Data
+
+
+def test_attach_direct(pal):
+ data = b'some data'
+ pal.attachments['direct.txt'] = data
+ assert pal.attachments['direct.txt'].get_file().read_bytes() == data
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_errors.py
new/pikepdf-8.10.1/tests/test_errors.py
--- old/pikepdf-8.9.0/tests/test_errors.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_errors.py 2023-12-17 10:31:00.000000000
+0100
@@ -46,7 +46,7 @@
def test_system_error():
with pytest.raises(FileNotFoundError):
- pikepdf._core._test_file_not_found()
+ pikepdf._core._test.fopen_nonexistent_file()
@skip_if_pypy
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_formxobject.py
new/pikepdf-8.10.1/tests/test_formxobject.py
--- old/pikepdf-8.9.0/tests/test_formxobject.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_formxobject.py 2023-12-17
10:31:00.000000000 +0100
@@ -3,7 +3,7 @@
from __future__ import annotations
-from pikepdf import Dictionary, Name, Object, Pdf, Stream
+from pikepdf import Dictionary, Name, Object, Page, Pdf, Stream
# pylint: disable=e1137
@@ -30,7 +30,7 @@
image = Stream(pdf, image_data)
image.stream_dict = Object.parse(
- """
+ b"""
<<
/Type /XObject
/Subtype /Image
@@ -71,13 +71,15 @@
contents = Stream(pdf, stream)
- page = pdf.make_indirect(
- {
- '/Type': Name('/Page'),
- '/MediaBox': mediabox,
- '/Contents': contents,
- '/Resources': resources,
- }
+ page = Page(
+ pdf.make_indirect(
+ {
+ '/Type': Name('/Page'),
+ '/MediaBox': mediabox,
+ '/Contents': contents,
+ '/Resources': resources,
+ }
+ )
)
pdf.pages.append(page)
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_io.py
new/pikepdf-8.10.1/tests/test_io.py
--- old/pikepdf-8.9.0/tests/test_io.py 2023-12-11 00:24:59.000000000 +0100
+++ new/pikepdf-8.10.1/tests/test_io.py 2023-12-17 10:31:00.000000000 +0100
@@ -197,7 +197,7 @@
def test_logging(caplog):
caplog.set_level(logging.INFO)
- pikepdf._core._log_info("test log message")
+ pikepdf._core._test.log_info("test log message")
assert [("pikepdf._core", logging.INFO)] == [
(rec[0], rec[1]) for rec in caplog.record_tuples
]
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_matrix.py
new/pikepdf-8.10.1/tests/test_matrix.py
--- old/pikepdf-8.9.0/tests/test_matrix.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_matrix.py 2023-12-17 10:31:00.000000000
+0100
@@ -11,6 +11,7 @@
from pikepdf import Array
from pikepdf._core import Matrix, Rectangle
from pikepdf.models import PdfMatrix
+from pikepdf.objects import Dictionary
def allclose(m1, m2, abs_tol=1e-6):
@@ -69,6 +70,21 @@
def test_default_is_identity(self):
assert Matrix() == Matrix(1, 0, 0, 1, 0, 0)
+ def test_not_enough_args(self):
+ with pytest.raises(TypeError):
+ Matrix(1, 2, 3, 4, 5)
+
+ def test_tuple(self):
+ assert Matrix() == Matrix((1, 0, 0, 1, 0, 0))
+ with pytest.raises(ValueError):
+ Matrix((1, 2, 3, 4, 5))
+
+ def test_failed_object_conversion(self):
+ with pytest.raises(ValueError):
+ assert Matrix(Array([1, 2, 3]))
+ with pytest.raises(ValueError):
+ assert Matrix(Dictionary(Foo=1))
+
def test_accessors(self):
m = Matrix(1, 2, 3, 4, 5, 6)
assert m.a == 1
@@ -145,3 +161,6 @@
assert (0, 0) < m.transform((1, 0)) < (1, 1)
m = Matrix().rotated(-45)
assert (0, 0) < m.transform((1, 0)) < (1, -1)
+
+ def test_latex(self):
+ assert '\\begin' in Matrix(1, 0, 0, 1, 0, 0)._repr_latex_()
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_object.py
new/pikepdf-8.10.1/tests/test_object.py
--- old/pikepdf-8.9.0/tests/test_object.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_object.py 2023-12-17 10:31:00.000000000
+0100
@@ -112,14 +112,14 @@
except InvalidOperation:
return # PDF doesn't support exponential notation
try:
- py_d = Object.parse(decstr)
+ py_d = Object.parse(decstr.encode('pdfdoc'))
except RuntimeError as e:
if 'overflow' in str(e) or 'underflow' in str(e):
- py_d = Object.parse(str(f))
+ py_d = Object.parse(f.encode('pdfdoc'))
assert isclose(py_d, d, abs_tol=1e-5), (d, f.hex())
else:
- with pytest.raises(PdfError):
+ with pytest.raises(PdfError), pytest.deprecated_call():
Object.parse(str(d))
@@ -325,11 +325,11 @@
assert Name('/Foo') != String('/Foo')
def test_numbers(self):
- self.check(Object.parse('1.0'), 1)
- self.check(Object.parse('42'), 42)
+ self.check(Object.parse(b'1.0'), 1)
+ self.check(Object.parse(b'42'), 42)
def test_bool_comparison(self):
- self.check(Object.parse('0.0'), False)
+ self.check(Object.parse(b'0.0'), False)
self.check(True, 1)
def test_string(self):
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_page.py
new/pikepdf-8.10.1/tests/test_page.py
--- old/pikepdf-8.9.0/tests/test_page.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_page.py 2023-12-17 10:31:00.000000000
+0100
@@ -130,14 +130,16 @@
def test_add_unowned_page(): # issue 174
pdf = Pdf.new()
d = Dictionary(Type=Name.Page)
- pdf.pages.append(d)
+ pdf.pages.append(Page(d))
def test_failed_add_page_cleanup():
pdf = Pdf.new()
d = Dictionary(Type=Name.NotAPage)
num_objects = len(pdf.objects)
- with pytest.raises(TypeError, match="only pages can be inserted"):
+ with pytest.raises(
+ TypeError, match="only pages can be inserted"
+ ), pytest.deprecated_call():
pdf.pages.append(d)
assert len(pdf.pages) == 0
@@ -148,7 +150,9 @@
# But we'd better not delete an existing object...
d2 = pdf.make_indirect(Dictionary(Type=Name.StillNotAPage))
- with pytest.raises(TypeError, match="only pages can be inserted"):
+ with pytest.raises(
+ TypeError, match="only pages can be inserted"
+ ), pytest.deprecated_call():
pdf.pages.append(d2)
assert len(pdf.pages) == 0
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_pages.py
new/pikepdf-8.10.1/tests/test_pages.py
--- old/pikepdf-8.9.0/tests/test_pages.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_pages.py 2023-12-17 10:31:00.000000000
+0100
@@ -202,11 +202,10 @@
pdf.pages[0::2] = pdf2.pages[0:1]
-@pytest.mark.timeout(1)
def test_self_extend(fourpages):
pdf = fourpages
- with pytest.raises(ValueError, match="source page list modified during
iteration"):
- pdf.pages.extend(pdf.pages)
+ pdf.pages.extend(pdf.pages)
+ assert len(pdf.pages) == 8
def test_one_based_pages(fourpages):
@@ -233,7 +232,7 @@
pdf = fourpages
with pytest.raises(TypeError):
pdf.pages.insert(0, 'this is a string not a page')
- with pytest.raises(TypeError):
+ with pytest.raises(TypeError), pytest.deprecated_call():
pdf.pages.insert(0, Dictionary(Type=Name.NotAPage, Value="Not a page"))
@@ -361,10 +360,20 @@
fourpages.pages.remove(p=-1)
+def test_remove_by_ref(fourpages):
+ second_page = fourpages.pages[1]
+ assert second_page == fourpages.pages[1]
+ fourpages.pages.remove(second_page)
+ assert second_page not in fourpages.pages
+ assert len(fourpages.pages) == 3
+ with pytest.raises(ValueError):
+ fourpages.pages.remove(second_page)
+
+
def test_pages_wrong_type(fourpages):
- with pytest.raises(TypeError):
+ with pytest.raises(TypeError), pytest.deprecated_call():
fourpages.pages.insert(3, {})
- with pytest.raises(TypeError):
+ with pytest.raises(TypeError), pytest.deprecated_call():
fourpages.pages.insert(3, Array([42]))
@@ -478,7 +487,7 @@
p = Pdf.new()
d = Dictionary(Type=Name.Page, MediaBox=[0, 0, 612, 792],
Resources=Dictionary())
for n in range(5):
- p.pages.append(d)
+ p.pages.append(Page(d))
p.pages[n].Contents = Stream(p, b"BT (Page %s) Tj ET" %
str(n).encode())
p.Root.PageLabels = p.make_indirect(
@@ -529,3 +538,11 @@
)
with pytest.raises(ValueError):
graph.pages.from_objgen(graph.pages[0].Contents.objgen)
+
+
+def test_page_iteration(graph, fourpages):
+ fourpages_iter = iter(fourpages.pages)
+ next(fourpages_iter) # Discard
+ next(fourpages_iter) # Discard
+ graph.pages.extend(fourpages_iter) # Append remaining two
+ assert len(graph.pages) == 3
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/pikepdf-8.9.0/tests/test_sanity.py
new/pikepdf-8.10.1/tests/test_sanity.py
--- old/pikepdf-8.9.0/tests/test_sanity.py 2023-12-11 00:24:59.000000000
+0100
+++ new/pikepdf-8.10.1/tests/test_sanity.py 2023-12-17 10:31:00.000000000
+0100
@@ -23,7 +23,7 @@
from packaging.version import Version
import pikepdf
-from pikepdf import Name, Object, Pdf, Stream
+from pikepdf import Name, Object, Page, Pdf, Stream
def test_minimum_qpdf_version():
@@ -106,7 +106,7 @@
'/Resources': resources,
}
qpdf_page_dict = page_dict
- page = pdf.make_indirect(qpdf_page_dict)
+ page = Page(pdf.make_indirect(qpdf_page_dict))
pdf.pages.append(page)
pdf.save(outdir / 'hi.pdf')
