Commit Graph

448 Commits

Author SHA1 Message Date
5c50e5b378 Embryo of code for openMP parallelization of LCS alignment but
deactivated for now because can't make it compile with cython/clang
2016-12-20 11:46:58 +01:00
9c71b06117 Removed deprecated TODOs 2016-12-19 14:36:40 +01:00
3bf5260174 Merge branch 'master' of git@git.metabarcoding.org:obitools/obitools3.git 2016-12-19 10:31:18 +01:00
857a5198e4 Updated `obi lcs` for the LCS alignment of two columns 2016-12-16 19:40:36 +01:00
d99447c12b C function for LCS alignment of two columns, and optimized and fixed
line count bug in function to align one column
2016-12-16 19:39:02 +01:00
303bd6f445 Added function to build kmer table for 2 columns, and fixed bug (with
line count) when building kmer table of one column
2016-12-16 19:10:18 +01:00
490f5fe6b9 Updated deprecated code in cython API for columns (using line count of
view instead of column)
2016-12-16 19:04:21 +01:00
191c83aafc Added missing *.cfiles 2016-12-15 15:28:34 +01:00
9b24818fe2 Refactored alignment code for minimum redundancy between the function
that aligns 1 column and the function that aligns 2 columns
2016-12-13 17:18:12 +01:00
fc805e5443 Remove some warnings in the editor 2016-12-13 08:29:22 +01:00
8d7ef7d3d1 patch the distutils to add the C source directory in the include path.
This should solve most of the compilation problems related to .h files
located in this directory
2016-12-13 08:02:09 +01:00
8afb1644e9 Alignment: API rework. 'obi align' is now 'obi lcs', and the results are
now written to columns automatically created in the output view, all
optimally handled at the C level.
2016-12-12 11:58:59 +01:00
fa4e4ffaff Changed the cython API to create new views so as to have different
functions for the different cases
2016-12-07 14:17:57 +01:00
936be64c34 Goes with 5e0c9f87 (missing ';' and fixed compilation warnings) 2016-12-05 11:18:29 +01:00
5e0c9f878b Added the doc for the function building the element names, and a missing
free
2016-12-05 10:46:21 +01:00
852e5488c8 The default element names for columns with multiple elements per line
are now "O;1;2;...;n"
2016-12-02 17:54:51 +01:00
e60497651c Updated the documentation for the functions to set and get in the
context of a view
2016-11-30 12:22:47 +01:00
4ad8c16a73 Finished adding all the functions to directly set and get indices in
columns containing indices referring to any type of data.
2016-11-30 11:08:11 +01:00
6f6099687d Sequence alignment: if no sequence column is given and the view has the
type NUC_SEQS_VIEW, the default sequence column is aligned
2016-11-29 16:52:41 +01:00
98d0849653 Sequence alignment: added the possibility to specify the index of the
sequences to align in a column containing multiple sequences per line (C
level for now)
2016-11-29 16:15:02 +01:00
5fb025f310 When aligning, it is now quickly checked whether the sequences are
identical using their indexes
2016-11-28 11:39:29 +01:00
8ce6f6c80b Added an argument to specify whether the two sequences can be identical
when applying filters before aligning
2016-11-28 11:38:02 +01:00
3e53f9418b Added functions to recover the indexes themselves from any column
referring to indexed values
2016-11-28 11:35:19 +01:00
d40d2d0c76 Fixed error in documentation 2016-11-28 10:55:23 +01:00
f897e87600 When closing a view, it is now automatically checked that all OBI_QUAL
columns correspond to their associated OBI_SEQ column
2016-11-25 12:04:57 +01:00
70e056a2aa It is now impossible to open or clone a view that is not finished (= has
been closed at least once)
2016-11-24 11:19:07 +01:00
8abbfa203a Good file for commit 6fa9a8bd: When a view is cloned, a comment is added
to the new view specifying the name of the cloned view
2016-11-23 11:32:39 +01:00
6fa9a8bd76 When a view is cloned, a comment is added to the new view specifying the
name of the cloned view
2016-11-23 11:29:21 +01:00
76a4c6b14e Fixed a bug when cloning a view and checking its type 2016-11-23 11:28:17 +01:00
0ab9e6c05a When adding an existing column to a view, it is checked that the
column's line count is at least the view's line count. This can't be
more stringent for reasons that need to be rediscussed
2016-11-23 11:04:53 +01:00
70c49e214a Added the kmer filter to LCS alignments, and now obiblobs containing
encoded sequences are directly put in int16_t arrays for the alignment
2016-11-18 16:29:28 +01:00
08e67a090f Changed the inline functions syntax, which should make it compatible
with more compilers
2016-11-18 16:21:26 +01:00
621b4972db Functions to get obiblobs through views 2016-11-18 15:59:50 +01:00
7d022c1a52 If the indexer name is NULL when creating a column, it now becomes the
column name
2016-11-18 15:56:51 +01:00
1c71c195fc Goes with a0ebc2d8 2016-11-10 15:01:29 +01:00
54cfeffd85 Goes with 8f724f4f, forgotten file 2016-11-10 14:48:31 +01:00
a0ebc2d871 Functions to directly retrieve Obiblobs from indexers 2016-11-10 14:45:28 +01:00
8f724f4f8e Some code refactoring 2016-11-09 16:48:00 +01:00
359578814b Added view type property to OBIView cython class and updated obi export
to use it
2016-11-08 17:49:59 +01:00
51b23915ca Added properties for Nuc_Seq cython classes (and updated commands using
them)
2016-11-08 16:59:32 +01:00
b5b889c4a2 Fixed the OBI_Nuc_Seq_Stored cython class not being up to date with the
new properties of its parent class
2016-11-08 11:26:37 +01:00
36ac315125 Fixed bugs with python view type when creating a new view, and a bug
when trying to guess the obi type of a nucleotide sequence when its type
was bytes
2016-11-08 11:23:54 +01:00
8291693309 obi grep: updated to work with the new line selection class and within
the local sequence environment, and progress bar functioning
2016-11-08 11:19:12 +01:00
4bc19c3e49 obi export: view type is now checked and progress bar functioning 2016-11-08 11:17:20 +01:00
2d2fe5279d Added functions to add new taxa to a taxonomy with handling of
associated *.ldx files
2016-11-03 17:59:21 +01:00
2504bf0fa9 Added an iterator to the OBI_Taxonomy cython class 2016-11-02 11:08:18 +01:00
d8a257e711 Taxonomy handling functions in C. Features: read taxdump, read binary
files, write binary files. Not fully handled yet: *.adx, *.pdx, *.ldx,
merged.dmp and delnodes.dmp files.
2016-10-27 18:56:11 +02:00
b63d0fb9fb Added C functions to write .rdx, .tdx, .ndx binary taxonomy files from a
taxonomy C structure
2016-10-14 17:03:10 +02:00
0dfd67ec89 The endianness of binary taxonomy files is now correctly checked 2016-10-10 17:04:29 +02:00
0faaac49cf The taxonomy directory of the DMS is now automatically created with the
DMS
2016-10-10 17:02:51 +02:00