tools: remove openfst check #1531
Conversation
KarelVesely84
commented
Apr 4, 2017
•
KarelVesely84
commented
- we will use the local open-fst, while we still can have the system-wide open-fst (not used by kaldi)
- (note: this setup was previously forbidden)
* Now we will use the local open-fst, while we still can have the system-wide open-fst (this setup was previously forbidden).
|
I better double-checked that this will cause no problems... OPENFSTINC = /mnt/matylda5/iveselyk/DEVEL/kaldi-official/tools/openfst/include The headers are okay, as the '-I${OPENFSTINC}' option has priority over the system includes. The related documentation is here: Headers [4th paragraph]: https://gcc.gnu.org/onlinedocs/cpp/Search-Path.html#Search-Path |
|
But I believe the OpenFst headers would be installed to a system location,
and system-level includes always take precedence over local ones.
…On Tue, Apr 4, 2017 at 8:06 AM, Karel Vesely ***@***.***> wrote:
I better double-checked that this causes no problems...
In `kaldi.mk' there are the absolute paths to the local OpenFST:
OPENFSTINC = /mnt/matylda5/iveselyk/DEVEL/kaldi-official/tools/openfst/
include
OPENFSTLIBS = /mnt/matylda5/iveselyk/DEVEL/kaldi-official/tools/openfst/
lib/libfst.so
OPENFSTLDFLAGS = -Wl,-rpath=/mnt/matylda5/iveselyk/DEVEL/kaldi-official/
tools/openfst/lib
The headers are okay, as the '-I${OPENFSTINC}' option has priority over
the system dirs.
The library is also okay, as the LD_LIBRARY_PATH is stored to the binary
by the '-rpath' options.
The related documentation is here:
Headers [4th paragraph]: https://gcc.gnu.org/onlinedocs/cpp/Search-Path.
html#Search-Path
Library [section 3.4]: http://tldp.org/HOWTO/Program-Library-HOWTO/shared-
libraries.html
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#1531 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/ADJVu22E7oY0_JNZLVOp0nX8a8gqWgNpks5rslyGgaJpZM4My1mv>
.
|
|
Hi apparently this is not true (It's from the GCC docs):
The order for
|
|
hm, OK. Merging. |
|
Thanks Dan! |
* 'master' of https://github.com/kaldi-asr/kaldi: [src] Cosmetic change: remove 'train.tra' from usage messages (kaldi-asr#1529) [src] cudamatrix: speed up AddColSumMat with transfrom reduce kernel template (kaldi-asr#1530) [build]: remove openfst check (kaldi-asr#1531) [build,src,doc] Modify get_version.sh to deal better with whitespace (avoid space in version); minor fixes (kaldi-asr#1526) [scripts,egs] Adding options for using PCA instead of LDA+MLLT for ivectors used in ASR. Results are reported in the default TDNN recipe in AMI. Updating steps/online/nnet2/{train_diag_ubm.sh,train_ivector_extractor.sh} so that they now backup the contents of their destination directory if it already exists. (kaldi-asr#1514) [src] (minor) Added missing SetZero() to NaturalGradientAffineComponent::Scale() if scale==0.0 (kaldi-asr#1522) [src,doc] Fix several unrelated minor problems. Thanks: gaoxinglong [src] Adding noexcept to hashing function objects (kaldi-asr#1519) [egs] Fix to egs/wsj/s5/run.sh (unset variable) (kaldi-asr#1517) [misc] remove eXecute permissions where not needed (kaldi-asr#1515) [src,scripts]: Several unrelated cosmetic changes [egs] fixes to babel pipeline; thanks to Fred Richardson (kaldi-asr#1509) [src] Fix exit code of extract-rows.cc (kaldi-asr#1510)
It appears there may be no good reason to disallow system-wide OpenFst.
It appears there may be no good reason to disallow system-wide OpenFst.
