summaryrefslogtreecommitdiff
path: root/apt-pkg/contrib
AgeCommit message (Collapse)Author
2016-05-01Don't copy strings in Startswith, EndswithAdrian Wielgosik
2016-04-28FileFd: avoid further writing if file failedDavid Kalnischkies
If the file is in a failed state there is no point in trying to flush out the buffers as the file is to be discarded anyhow & its likely all this flushing is producing is additional error messages. Git-Dch: Ignore
2016-04-25properly format multiline error messagesDavid Kalnischkies
2016-04-03use buffered writing for InRelease splittingDavid Kalnischkies
Hardly noticeable, but given that we have the option to easily enable it, lets enable it as every newline in the message is written individually by the code.
2016-04-03consistently add APT_OVERRIDE as method markerDavid Kalnischkies
Some methods had it missing, some used the keyword directly, which isn't a problem as it is a cc file, but for consistency lets stick to our macro for now. Git-Dch: Ignore
2016-03-28Allow lowering trust level of a hash via configJulian Andres Klode
Introduces APT::Hashes::<NAME> with entries Untrusted and Weak which can be set to true to cause the hash to be treated as untrusted and/or weak.
2016-03-24Set hidden visibility for StringViewJulian Andres Klode
This avoids templates using StringView to be exported, such as std::vector<StringView*>::emplace_back(). Gbp-Dch: ignore
2016-03-13Do not consider SHA1 usableJulian Andres Klode
SHA1 is not reasonably secure anymore, so we should not consider it usable anymore. The test suite is adjusted to account for this.
2016-03-07Fix several typosVeres Lajos
This effectively merges branch 'typofixes-vlajos-20150807' of github.com:vlajos/apt with the following commit: commit 13cacb3e2e2352ba701e769fc889e3344fabbf7e Author: Veres Lajos <vlajos@gmail.com> Date: Sun Aug 9 00:12:53 2015 +0100 typofix - https://github.com/vlajos/misspell_fixer It has been rebased for a better commit message.
2016-03-06Fix lzma write support to handle "try again" caseColin Watson
The liblzma-based write code needs the same tweaks that the read code already has to cope with the situation where lzma_code returns zero the first time through because avail_out is zero, but will do more work if called again. This ports the read tweaks to the write code as closely as possible (including matching comments etc.). Closes: #751688
2016-02-10Set seekpos to 0 after reopening files in Seek()Julian Andres Klode
If we just reopened the file, we also need to reset the current seek position when we reset the buffer, otherwise the code will not try to seek to the position given to Skip (from 0), but will try to seek to old offset + the position given to skip. Closes: #812994, #813000
2016-02-04Use some semantically more correct buffer operationsJulian Andres Klode
When writing into the buffer write to free() bytes starting at getend(), instead of buffersize_max bytes at get() -> get() is a read pointer. This makes no difference in practice though, as we reset the buffer before the call, so start = end = 0. Gbp-Dch: ignore
2016-02-04Correctly report write errors when flushing buffered writerJulian Andres Klode
We cannot just return false without setting an error, as InternalWrite does not set one itself.
2016-02-03Avoid temporary strings in SubstVar.Adrian Wielgosik
Microoptimization, but still gives a measurable 2-3% improvement when using commands with lots of output like `apt list`.
2016-02-01Do not buffer writes larger than the buffer if possibleJulian Andres Klode
It makes no sense to split a large block into multiple small blocks, so when we have the chance to write them unbuffered, do so.
2016-02-01BufferedWriteFileFdPrivate: Simplify InternalWrite()Julian Andres Klode
We do not need the loop, FileFd::Private() handles this for us. Gbp-Dch: ignore
2016-02-01simple_buffer::write: Use free() instead of maxsize - size()Julian Andres Klode
We want to check whether the amount of free space is smaller than the requested write size. Checking maxsize - size() is incorrect for bufferstart >= 0, as size() = end - start. Gbp-Dch: ignore
2016-01-30fix "Mismatched free() / delete / delete []" in simple_bufferStefan Bühler
2016-01-26drop explicit check for EWOULDBLOCK if it has the same value as EAGAINDavid Kalnischkies
gcc correctly reports that we check for the same value twice, expect that the manpage of read(2) tells us to do it for portability, so to make both sides happy lets add a little #if'ing here. Reported-By: gcc-6 Git-Dch: Ignore
2016-01-15string_view: Drop constexpr constructor for standard compatibilityJulian Andres Klode
APT::StringView is supposed to be a temporary measure, until support for the standardized string_view is widely available. Introducing additional unstandardized features just makes porting to the standard version harder. The constexpr constructor also won't have any real effect on most systems, as the compiler will happily optimise the strlen() call away for constant strings. Gbp-Dch: ignore
2016-01-15provide a constexpr char[] overload for APT::StringViewDavid Kalnischkies
The commit also adds a few trivial tests Git-Dch: Ignore
2016-01-15return correct position in APT::StringView::(r)findDavid Kalnischkies
The position returned is supposed to be the position of the character counted from the start of the string, but if we used the substr calling overloads the skipped over prefix wasn't considered. The pos parameter of rfind had also the wrong semantic.
2016-01-08Store the size of strings in the cacheJulian Andres Klode
By storing the size of the string in the cache, we can make use of it when comparing the names in the hashtable in pkgCache::FindGrp.
2016-01-08HashSumValue::Set: Do not provide const char* overloadJulian Andres Klode
Hide the std::string overload instead of providing a const char * one, the old variant was stupid. Gbp-Dch: ignore
2016-01-08operator==(char*, StringView) use StringView.operator==Julian Andres Klode
Use the same path for both comparisons, as the operator== path is faster than just calling compare() - it avoids any comparison if the size differs. Gbp-Dch: ignore
2016-01-07StringView: rfind: pos should be end of substr, not startJulian Andres Klode
Gbp-Dch: ignore
2016-01-07StringView: pos argument default should be nposJulian Andres Klode
Gbp-Dch: ignore
2016-01-07StringView::rfind(): Call rfind() instead of find() on substJulian Andres Klode
Thanks: Niels Thykier for reporting on IRC Gbp-Dch: ignore
2016-01-07Switch performance critical code to use APT::StringViewJulian Andres Klode
This improves performance of the cache generation on my ARM platform (4x Cortex A15) by about 10% to 20% from 2.35-2.50 to 2.1 seconds.
2016-01-07Introduce internal APT::StringView classJulian Andres Klode
The class APT::StringView implements a drop-in replacement for a subset of C++17 std::string_view() features. It will be dropped at a later point and may not be used in public interfaces.
2016-01-07CopyFile: Use 64 * 1024 instead of 64000 as buffer sizeJulian Andres Klode
This is a multiple of the page size and thus results in less page faults, speeding up copying. Also, while we're at at, unify all uses of that size in a constant variable APT_BUFFER_SIZE.
2016-01-07FileFd: (native) LZ4 supportJulian Andres Klode
Implement native support for LZ4 compression, using the official lz4 library.
2016-01-03simple_compressor: Provide some accessors for end and freeJulian Andres Klode
This makes code easier to read, and somewhat more correct. Gbp-Dch: ignore
2016-01-03simple_buffer: Allow buffer size to changeJulian Andres Klode
Gbp-Dch: ignore
2015-12-29strutl.cc: Add declarations for the compat _ascii() functionsJulian Andres Klode
This shuts up gcc Gbp-Dch: ignore
2015-12-29Turn tolower_ascii() and isspace_ascii() into inline functionsJulian Andres Klode
To preserve compatibility, the new inline functions have _inline as a suffix, and a macro defines the old names to refer to the inline variants. The old functions are still preserved for binary compatibility. Also simplify the implementation of both functions.
2015-12-28BufferedFileFdPrivate: Make InternalFlush() save against errorsJulian Andres Klode
Previously, if flush errored inside the loop, data could have already been written to the wrapped descriptor without having been removed from the buffer. Also try to work around EINTR here. A better solution might be to have the individual privates detect an interrupt and return 0 in such a case, instead of relying on errno being untouched in between the syscall and the return from InternalWrite.
2015-12-28BufferedWriter: flushing: Check for written < size instead of <=Julian Andres Klode
This avoids some issues with InternalWrite returning 0 because it just cannot write stuff at the moment.
2015-12-27FileFd: Add a buffered writing modeJulian Andres Klode
This is somewhat experimental right now, and might not work for everyone, so it is on an opt-in basis.
2015-12-27FildFd: Introduce a Flush() function and call it from Close()Julian Andres Klode
The flush function can be used for buffered writers.
2015-12-27FileFdPrivate: Add getter and setter for fieldsJulian Andres Klode
We will soon implement a buffered writing decorator and we will need to forward attribute changes to those.
2015-12-27fileutl: simple_buffer: Add write() and full() methodsJulian Andres Klode
These can be used to implement write buffering Gbp-Dch: ignore
2015-12-27fileutl: simple_buffer: Mark accessors as constJulian Andres Klode
Suggested by David. Gbp-Dch: ignore
2015-12-27FileFdPrivate: Extract SimpleBuffer and mark it as hiddenJulian Andres Klode
Gbp-Dch: ignore
2015-12-27Hex2Digit: Do not use isxdigit()Niels Thykier
We directly check if we are a hex digit in HexDigit, so use that information. [jak@debian.org: Commit message wording]
2015-12-27Convert most callers of isspace() to isspace_ascii()Julian Andres Klode
This converts all callers that read machine-generated data, callers that might work with user input are not converted.
2015-12-27Introduce isspace_ascii() for use by parsersJulian Andres Klode
This is like isspace(), but ignores the current locale.
2015-12-26Refactor InternalReadLine to not unroll Size == 0 caseJulian Andres Klode
There is not much point and this is more readable. Gbp-Dch: ignore
2015-12-26Change InternalReadLine to always use buffer.read() return valueJulian Andres Klode
This is mostly a documentation issue, as the size we want to read is always less than or equal to the size of the buffer, so the return value will be the same as the size argument. Nonetheless, people wondered about it, and it seems clearer to just always use the return value.
2015-12-26Get rid of memmove() in our read bufferingJulian Andres Klode
This further improves our performance, and rred on uncompressed files now spents 78% of its time in writing. Which means that we should really look at buffering those.