Commit Graph

3148 Commits

Author SHA1 Message Date
Ward Fisher
0eff434098 Renamed test file to be more reflective for what it is doing. 2016-02-16 14:54:10 -07:00
Ward Fisher
4ccad07a6b Have duplicated issue reported in #221 at github. 2016-02-16 14:52:03 -07:00
Ward Fisher
654a6e39e3 Additional work debugging issue. 2016-02-16 14:40:09 -07:00
Ward Fisher
1807cb7aa2 Roughing in a test to create the failure described in https://github.com/Unidata/netcdf-c/issues/221 2016-02-16 11:00:09 -07:00
Ward Fisher
47dcd00bb7 Added test file for debugging https://github.com/Unidata/netcdf-c/issues/221. 2016-02-16 10:23:53 -07:00
Ward Fisher
7e7fa31b5f Merge pull request #215 from gsjaardema/patch-5
Support user-specified hdf5 and hdf5_hl libraries
2016-02-16 10:19:15 -07:00
Greg Sjaardema
ebc25610ef Merge branch 'master' into patch-5 2016-02-09 13:43:06 -07:00
Greg Sjaardema
126a319230 Fix version dependency.
Required PNetCDF version is 1.6.1 or greater; not 1.6.0 as I previously thought.

Note that this is pull request may (partially) satisfy Issue #194 which I just found.
2016-02-09 11:23:33 -07:00
Greg Sjaardema
d123157771 Enable PNetCDF settings in cmake build
One possible solution to finding the PNetCDF library and testing for the correct version in the cmake build.
2016-02-09 07:23:09 -07:00
Greg Sjaardema
edfffb4b14 Merge branch 'master' into patch-5 2016-02-09 07:05:54 -07:00
Ward Fisher
1a42d4ff99 Corrected version string to reflect development branch is 4.4.1 in support of correcting https://github.com/Unidata/netcdf-c/issues/214. 2016-02-08 17:36:04 -07:00
Greg Sjaardema
df58364f00 Support user-specified hdf5 and hdf5_hl libraries
If the netcdf cmake build is configured with explicitly specified hdf5 c and hl libraries:
```
    -DHDF5_C_LIBRARY:PATH=${ACCESS}/lib/libhdf5.dylib \
    -DHDF5_HL_LIBRARY:PATH=${ACCESS}/lib/libhdf5_hl.dylib \
    -DHDF5_INCLUDE_DIR:PATH=${ACCESS}/include \
```
Then it looks like the HDF5_C_LIBRARIES and HDF5_HL_LIBRARIES variables do not get set and then this causes the HDF5 libraries to be omitted from the `TLL_LIBS` symbol in liblib/CMakeLists.txt.  This is subsequently used to build `ALL_TLL_LIBS` which is used for `LIBS` which is output as the `Extra libraries:        @LIBS@` in libnetcdf.settings.in 

The output for `-- Linking against:` is also incorrect as it omits the hdf5 libraries:   Output without suggested change:
```
-- Linking against:      /usr/lib/libdl.dylib;/usr/lib/libm.dylib;/opt/local/lib/libz.dylib;/opt/local/lib/libcurl.dylib
(later in file)
Extra libraries:	-ldl -lm -lz -lcurl
```
versus output with suggested change:
```
-- Linking against:      /Users/gdsjaar/src/seacas-parallel/lib/libhdf5_hl.dylib;/Users/gdsjaar/src/seacas-parallel/lib/libhdf5.dylib;/usr/lib/libdl.dylib;/usr/lib/libm.dylib;/opt/local/lib/libz.dylib;/opt/local/lib/libcurl.dylib
(later in file)
Extra libraries:	-lhdf5_hl -lhdf5 -ldl -lm -lz -lcurl
```
This also causes a link failure when building libnetcdf.dylib
2016-02-08 17:20:48 -07:00
Ward Fisher
55b24e2964 Added tst_mode to CMakeLists.txt 2016-02-04 19:12:15 +00:00
Ward Fisher
03b7a103ad Merge pull request #210 from Unidata/gh206
merging work @dmh did to address #206
2016-02-03 10:47:01 -07:00
Dennis Heimbigner
e03785069a Merge branch 'gh206' of https://github.com/Unidata/netcdf-c into gh206 2016-02-02 13:07:15 -07:00
Dennis Heimbigner
66fcf79d77 Github issue: https://github.com/Unidata/netcdf-c/issues/206
Re e-eupport VGQ-678069
It was noticed that an attribute value of "nan." was being treated as
legal (it should be "nan"). The reason is that sscanf was not be checked
to see that all the attribute value characters were being read.
Solution is to verify that all characters were being consumed.
2016-02-02 13:06:57 -07:00
Ward Fisher
20ce7edf32 Merge pull request #209 from Unidata/gh208
Fix github issue: https://github.com/Unidata/netcdf-c/issues/208
2016-02-02 12:42:00 -07:00
Dennis Heimbigner
80a07f2375 Merge branch 'gh206' of https://github.com/Unidata/netcdf-c into gh206 2016-02-01 16:21:32 -07:00
Dennis Heimbigner
b2a25bde13 Github issue: https://github.com/Unidata/netcdf-c/issues/206
Re e-eupport VGQ-678069
It was noticed that an attribute value of "nan." was being treated as
legal (it should be "nan"). The reason is that sscanf was not be checked
to see that all the attribute value characters were being read.
Solution is to verify that all characters were being consumed.
2016-02-01 16:19:34 -07:00
Dennis Heimbigner
45572f5971 Fix github issue: https://github.com/Unidata/netcdf-c/issues/208
Return an error when specifying deflation (compression) or fletcher32 on
a file created for parallel IO in netcdf-4.
2016-02-01 16:15:58 -07:00
Ward Fisher
990296c671 Merge pull request #205 from Unidata/hdf5init
Clean up the handling of hdf5 initialization
2016-01-29 11:07:54 -07:00
Ward Fisher
7c828bdb5e Merge branch 'master' into hdf5init 2016-01-29 10:31:52 -07:00
Ward Fisher
83df40b6f9 Merge pull request #204 from Unidata/contributing-toc
Added a table-of-contents to the contributing guideline file.
2016-01-29 10:12:43 -07:00
Dennis Heimbigner
d91ccdf54c Github issue: https://github.com/Unidata/netcdf-c/issues/206
Re e-eupport VGQ-678069
It was noticed that an attribute value of "nan." was being treated as
legal (it should be "nan"). The reason is that sscanf was not be checked
to see that all the attribute value characters were being read.
Solution is to verify that all characters were being consumed.
2016-01-28 22:18:51 -07:00
Dennis Heimbigner
b5ba424793 Clean up the handling of hdf5 initialization by
creating an nc4_hdf5_initialize(void) function
plus nc4_hdf5_initialized flag.
Also fix potential null exception in nc4internal.c
2016-01-28 16:19:38 -07:00
Ward Fisher
5b18f955dc Added a table-of-contents to the contributing guideline file. 2016-01-27 15:56:34 -07:00
Ward Fisher
f85d8a0ead Merge pull request #203 from Unidata/contributing
Add contributing guideline document based on `contributing.html` file added by @dmh.
2016-01-27 15:49:14 -07:00
Ward Fisher
93560f7555 Updated contributing markdown document. 2016-01-27 14:29:58 -07:00
Ward Fisher
bfb18e4f57 First draft of contributing.md 2016-01-26 16:06:19 -07:00
Ward Fisher
e81f5ed7d6 Adding a contributing guideline file. 2016-01-26 15:23:04 -07:00
Ward Fisher
427577be8d Merge branch 'gsjaardema-patch-6' into pull-consolidation 2016-01-26 10:10:54 -07:00
Ward Fisher
30e70df5b6 Merge branch 'gsjaardema-patch-5' into pull-consolidation 2016-01-26 10:09:51 -07:00
Ward Fisher
cf9a477daa Modified spacing in configure.ac. 2016-01-26 10:09:14 -07:00
Greg Sjaardema
e1696187ec Replace NC_MAX_DIMS with NC_MAX_VAR_DIMS
Same issue as in #197.
2016-01-25 14:55:29 -07:00
Greg Sjaardema
ba7d9c54bd Replace NC_MAX_DIMS with NC_TESTS_MAX_DIMS
Missed this file earlier somehow.  Same situation as fixed in fbc210f
2016-01-25 14:47:52 -07:00
Ward Fisher
fbc210f511 Alternative solution to https://github.com/Unidata/netcdf-c/pull/199 and https://github.com/Unidata/netcdf-c/pull/200. 2016-01-25 14:09:03 -07:00
Ward Fisher
f3cae2e6ce Merge branch 'master' of https://github.com/Unidata/netcdf-c 2016-01-25 11:29:12 -07:00
Ward Fisher
3769de2669 Merge branch 'gsjaardema-patch-2' 2016-01-25 11:10:32 -07:00
Ward Fisher
dc78c6a4a5 Merge pull request #197 from gsjaardema/patch-1
Replace NC_MAX_DIMS with correct NC_MAX_VAR_DIMS
2016-01-25 10:23:13 -07:00
Greg Sjaardema
04f1231f48 Fix incorrect array size specification
The count and start arrays are dimensioned to NC_MAX_DIMS even though the maximum size should be NDIMS which is set to 3; possibly, the maximum size could be 2 since only indices 0 and 1 are used to access both start and count.  I left it at NDIMS since that matches the number of items in the initialization and is consistent with other uses in the file.
2016-01-22 21:03:48 -07:00
Greg Sjaardema
e13d976e27 Replace NC_MAX_DIMS with correct NC_MAX_VAR_DIMS
The constant NC_MAX_DIMS is used to dimension the `dimids` arrays which are use to retrieve the dimension ids corresponding to the variables dimensions.  By definition, the maximum number of dimension ids is `NC_MAX_VAR_DIMS` as is documented in the `nc_inq_var` documentation:
http://www.unidata.ucar.edu/software/netcdf/netcdf-4/newdocs/netcdf-c/nc_005finq_005fvar.html

dimids   Returned vector of *ndimsp dimension IDs corresponding to the variable dimensions. The caller must allocate enough space for a vector of at least *ndimsp integers to be returned. The maximum possible number of dimensions for a variable is given by the predefined constant NC_MAX_VAR_DIMS. 

Although the default or standard values of NC_MAX_DIMS and NC_MAX_VAR_DIMS are the same, the correct value to be used here is NC_MAX_VAR_DIMS and not NC_MAX_DIMS.  Even though this currently works, it could fail if either NC_MAX_VAR_DIMS or NC_MAX_DIMS is changed and it can also provide an incorrect function usage example that may mislead developers trying to determine the correct usage of the function.
2016-01-22 20:54:52 -07:00
Ward Fisher
0ee9a4f59e Further cleanup of generated files. 2016-01-22 12:04:43 -07:00
Ward Fisher
79c021712a Removed generated tests from source control. Ensuring generated files are regenerated if m4 is present. 2016-01-22 12:03:28 -07:00
Ward Fisher
a3ec3a5d94 Merge pull request #193 from Unidata/doc-update
Doc update
2016-01-21 13:43:06 -07:00
Ward Fisher
8d162e9097 Merge branch 'master' into doc-update 2016-01-21 12:39:57 -07:00
ddirks
c8e922dc92 Tweaks to css
Fix placement of version selection dropdown and search box.
2016-01-21 11:19:23 -07:00
Ward Fisher
1805ef0668 Tweaks to style sheet with assistance from @ddirks 2016-01-21 10:11:17 -07:00
Ward Fisher
548e279fbe Merge branch 'doc-update' of https://github.com/Unidata/netcdf-c into doc-update 2016-01-19 16:24:59 -07:00
Ward Fisher
97883f81d5 Adding an obsolete fan_utils file. 2016-01-19 16:22:57 -07:00
Ward Fisher
94451d18d8 Adding fan_utils.html. 2016-01-19 16:15:51 -07:00