17605996d6
remove whitespace before EOL
2015-11-02 17:24:07 +01:00
e0efc4b91f
Re-order headers in modified files
2015-10-20 15:46:09 -07:00
741fd1e941
Remove unused define
2015-10-20 15:34:54 -07:00
081201d6e8
Remove excess includes, switch to C++ includes
2015-10-18 13:10:07 -07:00
a6ddaed4a5
avoid duplicates in job queues for pending ways and relations
...
If the pending tracker of output was ahead of the pending tracker
of middle when merging the two then an id coming from the middle
tracker was unconditionally added even if it was present in the
output tracker and therefore added from there.
Fixes #419
2015-07-31 21:40:19 +02:00
3d44d34db1
Move db options into database_options_t
2015-07-25 19:48:12 -07:00
8eecf7fcf6
Rename tag to tag_t
2015-07-20 19:46:46 -07:00
1aed75474e
Replace wkt_itr with foreach loops and auto
2015-07-19 13:46:47 -07:00
55922d6ebc
Replace BOOST_FOREACH with new C++11 for loops
2015-07-18 10:06:01 -07:00
8b0ffe4a6a
Replace boost::shared_ptr with std::shared_ptr
2015-07-18 02:14:31 -07:00
6b5a0fffbf
Move from NULL to nullptr
...
This leaves alone the o5m code, geos, and libosmium.
2015-07-17 16:30:35 -07:00
25bd886ed3
Update middle return types from int
...
g++ was warning of a signed/unsigned conversion
2015-07-16 16:13:47 -07:00
588431536f
remove scale everywhere but in ramNode
2015-05-30 15:46:55 +02:00
72503fc2f7
Merge commit '4a5fc188e70a24c9772765c5acbb324e18b07be0' into static_analysis_2
...
Picks off parts of #299
2015-04-22 13:00:25 -07:00
275339a811
second attempt to fix broken way_area assignment
2015-04-13 23:31:06 +02:00
09da452026
correctly overwrite way_area for each polygon part
2015-04-13 21:40:21 +02:00
15f0d9802a
reenable Wall and fix resulting warnings
2015-04-13 20:55:24 +02:00
8d803dc4d7
new types for keyval and for tag, member and node lists
...
Replace all hand-crafted lists by a simple std::vector. Make data
from the parser const and copy convered data in new vectors.
Coding style normalized in code touched by the change.
2015-04-12 21:59:36 +02:00
be29d2ebbb
Make use of boost::diagnostic_information compatible with older versions of Boost.
2015-02-22 10:41:36 +00:00
3222fe1888
Catch exceptions in threads and print them upon join. This should stop thrown errors from aborting the thread without any message.
2015-02-22 09:40:17 +00:00
4a5fc188e7
Globally: catch by const ref
2015-02-21 15:15:26 +01:00
374e846874
';' outside function is C++11 extension
2015-02-21 12:59:12 +01:00
f1efe8753c
remove text-tree and make keyval a real C++ class
2014-12-28 21:38:47 +01:00
3d381ffe4f
replace keyval::initList with proper C++ constructor use
2014-12-25 17:53:11 +01:00
d13ce14519
Remove buffer.cpp
...
Fixes #200
2014-10-26 00:24:06 -07:00
c3a203fa5a
Clean up whitespace errors
2014-10-25 01:04:46 -07:00
9f0388b00a
fix some warinigs by removing unused variables
2014-10-02 16:50:38 +06:00
b81e9e5085
remove unused parameter names and GCC __attribute__
2014-10-02 12:55:49 +04:00
6debabccd2
dont set way done tracker when reprocessing relations as no one cares about it and it would be thread contention
2014-09-26 11:16:55 -04:00
1dfce321bc
fix a bug where threaded pending would not have access to the information about which ways were determined to be used in relations beforehand which would have lead to more ways in the lines table for example
2014-09-26 08:54:38 -04:00
756a77454a
dont add bogus ids to the pending threads' work queue
2014-09-25 11:01:01 -04:00
fa05c0735f
Merge remote-tracking branch 'origin/jh_threadpend' into jh_threadpend
...
Conflicts:
output-multi.cpp
output-pgsql.cpp
2014-09-24 17:47:10 -04:00
3f248b4e1b
Remove relation callbacks and replace with threaded relation processing.
2014-09-24 17:35:33 -04:00
0aa391a720
all tests passing with non global text_tree. but performance shouldnt be very good
2014-09-24 16:36:46 -04:00
5c86f4f679
Remove way callbacks which are no longer needed.
2014-09-24 15:27:41 -04:00
c50bf832ee
remove pending tracking from the ram middle processing as now the middle has all ways nodes relations in it
2014-09-23 16:49:36 -04:00
659b9bb706
refactored a bunch of native pointers to smart ones. tests still failing for relations need to check old branch to see exepcted pending counts
2014-09-23 15:11:39 -04:00
a8c5b9f916
remove some of the native pointers and finish up moving pending rels from threads back to main outputs
2014-09-23 11:57:28 -04:00
68ff8a8f58
change around output map to be a vector with the index in the vector being the id of the output so the threads can simply index into their vector of outputs
2014-09-23 09:17:13 -04:00
fe7ce22419
Populate pending job queue prior to cloning out the middles and outputs.
...
Some other bug fixes and refactor.
Most tests pass now, but three still fail.
2014-09-22 21:07:07 -04:00
28cfac3577
Use cloned middle in cloned outputs.
2014-09-22 09:25:16 -04:00
c0e4c99b74
Merge expires on stop in multi and pgsql outputs.
2014-09-19 17:21:44 -04:00
3b56428b98
Combine base classes way_cb_func and rel_cb_func into single base class cb_func. Move thread handling out of middle_t into osmtypes.cpp.
2014-09-17 07:55:44 -04:00
dec9b54d4d
Merge remote-tracking branch 'origin/mla_threadpend' into jh_threadpend
...
Conflicts:
.gitignore
2014-09-16 15:18:54 -04:00
fca5f04015
Add clone methods to output classes, which create clones using the copy constructors.
2014-09-15 17:06:49 -04:00
e47a48a2bf
Abstract expire_tiles output.
...
Rather than just outputting to a file at destructor time, which
has the potential to go horribly wrong if anything is thrown, this
now needs `output_and_destroy()` to be called before the object
goes out of scope.
The output is now effectively a functor object, so this allows it
to be swapped out for testing purposes.
2014-09-15 18:33:43 +01:00
75f3a75862
make MOAR copy constructors although because we are using pointer to parent class output_t we will need to change to having a pure virtual clone method.. to actually use this in the threads
2014-09-12 16:46:43 -04:00
95ab5d3f82
a performance step back but simplify the hand off between middle pending and backend pending. this should allow us to treat every job we push into the queue the same. after we've got it threaded we can go back and redesign it to query multiple ids at a time and do batches or whatever means we deem fit
2014-09-10 14:57:27 -04:00
eaedf6d1d7
add some small bookkeeping to id tracker to know how many marks it is carrying at any one time. clean out the rest of the db stuff that was there for pending
2014-09-05 11:52:32 -04:00
2565bf738f
since middle is not seprate from out we are now keeping all way tags in middle (ie sans filtering) that means that when an out gets at a way through the middle it cant expect its particular important tags to be the only ones there. therefore each out must filter using its own transform/export list before using a way
2014-06-17 11:28:22 -04:00
