Up until now aria2 file auto renaming worked by just appending a new
unique number to the file path, behind the file name extension, if any,
changing what most other programs consider the file extension in the
process.
Now, aria2 will attempt to insert the number before the file extension,
leaving the extension intact, so that e.g. a ".jpg" still is a ".jpg"
file and opens in your configured image viewer when clicking it.
If a file has no extension (incl. so called "dot files"), the number
will be appended to the file name as usual.
Note: This is a potentially breaking change that might break third party
scripts that rely on aria2 auto file renaming producing a certain format
for renamed files. Please fix your scripts accordingly.
Closes GH-709
Previously, we categorized options that can be used in
aria2.changeOption RPC method into 2 categories. The options in one
category can be applied on the fly, meaning that download continues to
be active while applying options. Another category includes options
which are only applicable when downloads are waiting or paused.
In this change, when active download is ordered to change options
which only applicable in waiting or paused state, it is now paused,
and then automatically restarted. Although we have limited number of
download concurrency, the pause and restart is done atomically, and
the download is inserted at the front of the queue, it is picked up
immediately if the concurrency regulation allows.
This option excludes seed only downloads when counting concurrent
active downloads (-j option). This means that if -j3 is given and
this option is turned on and 3 downloads are active and one of those
enters seed mode, then it is excluded from active download count (thus
it becomes 2), and the next download waiting in queue gets started.
But be aware that seeding item is still recognized as active download
in RPC method.
In my quest to explore the code and understand it, why not clean it up
in the progress. Most formatting provided by clang-format. ;)
Aside from formatting, also extracted some method and unnested some
control structs.
This change replaces the current 64 bit sequential GID with 64 bits
random bytes GID in an attempt to support persistent GID. Internally,
the GID is stored as uint64_t. For human representation and RPC
interface, GID is represented as 16 bytes hex string. For console
readout, 16 bytes are too long, so it is abbreviated to first 6 bytes.
When querying GID in RPC calls, user can speicfy the prefix of GID as
long as the prefix is shared by more than 1 GID entries.
The old implementation calculates download/upload statistics for a
RequestGroup by summing up all PeerStat objects. For global
statistics, those are summed together. This clearly incurs runtime
penalty and we introduced some kind of caching which updates
statistics every 250ms but it did not work right.
This change removes all these aggregation code, and instead makes
RequestGroup and RequestGroupMan objects hold NetStat object and
download/upload bytes are directly calculated by thier own NetStat.
This is far more simplar than the old way and less runtime penalty and
brings more accuracy.
DownloadHandlerConstants was simplified. MIME type handling in Accept
header was also reworked. DownloadContext's metalinkServerContacted_
is replaced with acceptMetalink_ and its boolean value is reverted.
RequestGroup and HttpRequest now do not hold vector of accepting
types. HttpRequest has the flag acceptMetalink_ which will be set by
the same value of DownloadContext::accpetMetalink_ and if it is true,
Metalink MIME types are added to Accept header field.
The only exception is zero-length file. If server tells file is
zero-length and --checksum option is given, aria2 now correctly checks
its checksum. There is one known issue: If downloaded file is
zero-length file and .aria2 file exists, it will not be deleted on
successful verification, because .aria2 file is not loaded.
The delay is caused because some Commands are only called in certain
interval(called refreshInterval, default, 1000ms). In aria2 download
stops when all Commands associated to it are stopped. Since some
Commands are called in each 1000ms by default, as mentioned before, we
have to wait for them. To fix this issue, we call
DownloadEngine::setRefreshInterval(0) when pausing/stopping downloads.
DownloadEngine::setRefreshInterval(0) makes refreshInterval 0 in one
shot.
When all segments are ignored, now DownloadFailureException is thrown.
And stop the download immediately. As described earlier, we call
DownloadEngine::setRefreshInterval(0) in catch block of
DownloadFailureException to eliminate delay.
Removed SharedHandle::isNull(). Instead we added operator* and
operator unspecified_bool_type. Removed use of WeakHandle and
replaced with raw pointer.
Fixed the bug that a file gets overwritten if -V is given and no
hash is provided. Fixed the bug that --dry-run leads download
error. Added RequestGroup::createCheckIntegrityEntry() which
correctly creates CheckIntegrityEntry objects and open files based
on -V option and the existence of control file.
* src/AbstractCommand.cc
* src/AbstractCommand.h
* src/ChecksumCheckIntegrityEntry.cc
* src/DownloadContext.cc
* src/DownloadContext.h
* src/FtpNegotiationCommand.cc
* src/HttpResponseCommand.cc
* src/PieceHashCheckIntegrityEntry.cc
* src/RequestGroup.cc
* src/RequestGroup.h
* src/RequestGroupEntry.cc
* src/RequestGroupEntry.h
Count the number of command used in HTTP(S)/FTP and the number of
connections in HTTP(S)/FTP independently. The former is used to
determin whether additional command is needed. The latter is used
to report user to how many connections are used in a download.
* src/AbstractCommand.cc
* src/AbstractCommand.h
* src/CreateRequestCommand.cc
* src/RequestGroup.cc
* src/RequestGroup.h
Hold reference to PieceStorage when paused in order to provide
progress information via XML-RPC.
* src/RequestGroup.cc
* src/RequestGroup.h
* src/RequestGroupMan.cc
Added aria2.pause and aria2.unpause XML-RPC method. aria2.pause
pauses the download denoted by gid. gid is of type string. The
status of paused download becomes "paused" and the download is
placed on the first position of waiting queue. As long as the
status is "paused", the download is not started. To change status
to "waiting", use aria2.unpause method. This method returns GID of
paused download. aria2.unpause changes the status of the download
denoted by gid from "paused" to "waiting". This makes the download
eligible to restart. gid is of type string. This method returns
GID of unpaused download.
* doc/aria2c.1.txt
* src/DownloadContext.cc
* src/FileEntry.cc
* src/OptionHandlerFactory.cc
* src/RequestGroup.cc
* src/RequestGroup.h
* src/RequestGroupMan.cc
* src/XmlRpcMethodFactory.cc
* src/XmlRpcMethodImpl.cc
* src/XmlRpcMethodImpl.h
* src/prefs.cc
* src/prefs.h
* src/usage_text.h
Added --always-resume and --max-resume-failure-tries option. If
--always-resume=false is given, when all given URIs does not
support resume or aria2 encounters N URIs which does not support
resume
(N is the value specified using --max-resume-failure-tries
option), aria2 download file from scratch. The default behavior
is --always-resume=true, which means if all URIs do not support
resume, download fails. I think this is OK because user normally
don't like to see that partially downloaded file is
overwritten(this is particularly true if file size is big). This
option is useful when aria2 is used as download backend and
graceful falling back to overwritten behavior is preferable.
Added exit status value 8, which means download failed because
server did not support resume.
* src/AbstractCommand.cc
* src/DefaultPieceStorage.cc
* src/DownloadCommand.cc
* src/DownloadResultCode.h
* src/FileEntry.h
* src/FtpNegotiationCommand.cc
* src/HttpResponse.cc
* src/HttpResponseCommand.cc
* src/OptionHandlerFactory.cc
* src/RequestGroup.cc
* src/RequestGroup.h
* src/SegmentMan.cc
* src/SegmentMan.h
* src/prefs.cc
* src/prefs.h
* src/usage_text.h
* test/DefaultPieceStorageTest.cc
* test/SegmentManTest.cc
Nils Maier