libtabula

Check-in [65286b8151]
Login

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:Changed all *.md to Unix line endings, and renamed one holdout *.txt file to *.md. The old Notepad excuse for using CRLF in these files no longer holds with *.md. There are still some *.txt files, but they're mostly CMakeLists.txt, which shouldn't be edited in Notepad to begin with, plus bmark.txt, which shouldn't be edited at all.
Timelines: family | ancestors | descendants | both | trunk
Files: files | file ages | folders
SHA1: 65286b81511d80f252d0ac780d74a0b7c9071336
User & Date: etr 2014-09-17 04:18:48
Context
2014-09-18
08:35
Ignoring any file or directory at the top level called "build*", so that "fossil clean" doesn't remove everything in the CMake build directory, if you have chosen to put it underneath the source root.

This establishes a convention, where multiple build directories must be named build*. For instance, a Windows user with multiple versions of VC++ installed might have "build-vs2008" and such. check-in: 44b78129b8 user: tangent tags: trunk

2014-09-17
04:18
Changed all *.md to Unix line endings, and renamed one holdout *.txt file to *.md. The old Notepad excuse for using CRLF in these files no longer holds with *.md. There are still some *.txt files, but they're mostly CMakeLists.txt, which shouldn't be edited in Notepad to begin with, plus bmark.txt, which shouldn't be edited at all. check-in: 65286b8151 user: etr tags: trunk
04:07
First rough cut at replacing all the Bakefile commentary in the MD files with CMake equivalents. check-in: d901ae7664 user: etr tags: trunk
Changes
Hide Diffs Unified Diffs Ignore Whitespace Patch

Changes to COPYING.md.

Changes to CREDITS.md.

Changes to ChangeLog.md.

Changes to HACKERS.md.

146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
...
166
167
168
169
170
171
172
173
174
175






176
177
178
179
180
181
182

* Perl and shell script rules are more or less the same
  as for C/C++, to the extent this makes sense.


**XML/HTML Dialects**

File types: `bkl`, `dbx`, `hta`

* Spaces for indents, size 2.  Shallow indents due to the
  high level of nesting occurring in such files, and spaces
  because they're not as annoying at shallow indent levels
  in editors that don't treat space indents like tabs.

* Unix line endings.  Again, these are intended to be viewed
................................................................................

File types: `md`, `txt`

* Spaces for indents, size 4.  Spaces because such files
  are often viewed in Notepad and similarly crippled text
  editors which use a default indent level of 8.

* DOS line endings, again for the Notepad reason.  And on
  modern Unixy platforms, the tools cope with DOS line endings
  reasonably well.  Better than the converse, anyway.







When in doubt, mimic what you see in the current code.  When still
in doubt, ask on the mailing list.


Testing Your Proposed Change
----







|







 







|
|
|
>
>
>
>
>
>







146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
...
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188

* Perl and shell script rules are more or less the same
  as for C/C++, to the extent this makes sense.


**XML/HTML Dialects**

File types: `dbx`, `hta`

* Spaces for indents, size 2.  Shallow indents due to the
  high level of nesting occurring in such files, and spaces
  because they're not as annoying at shallow indent levels
  in editors that don't treat space indents like tabs.

* Unix line endings.  Again, these are intended to be viewed
................................................................................

File types: `md`, `txt`

* Spaces for indents, size 4.  Spaces because such files
  are often viewed in Notepad and similarly crippled text
  editors which use a default indent level of 8.

* Unix line endings.  Notepad can't cope with LF line endings, and
  it is the default handler of `*.txt` on Windows, but then, such
  files are all build and test system files, so you really want to be
  editing them in a competent programmer's text editor to begin with,
  and such things understand LF line endings.

  As for `*.md`, there is no default association for that on Windows,
  so you might as well associate it with either a competent plain
  text editor or a Markdown editor/viewer.

When in doubt, mimic what you see in the current code.  When still
in doubt, ask on the mailing list.


Testing Your Proposed Change
----

Changes to INSTALL.md.

Changes to LICENSE.md.

Changes to README-Cygwin.md.

Changes to README-Linux.md.

Changes to README-Mac-OS-X.md.

Changes to README-Solaris.md.

Changes to README-examples.md.

Changes to README.md.

Name change from RELEASE-CHECKLIST.txt to RELEASE-CHECKLIST.md.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

38
39
40

41
42
43
44
45
- Change version number in configure.ac and libtabula.bkl.

  All other places the version number occurs are generated files
  created from one of these two.  If there's a corresponding *.in file
  for the one you're looking at, the version number was substituted in
  by autoconf from configure.ac.  Otherwise, the file was most likely
  created by the build system using the version number in libtabula.bkl.

- Run "make abicheck".  There should be no changes.

  You may have to run the following command in the current "stable"
  directory before this will succeed, since it depends on there being
  an ACC dump file in place already.

      $ abi-compliance-checker -lib tabula -dump abi.xml

  This assumes a working tree that looks something like this:

    projects/
	   libtabula/
	      4.0.0/                # unpacked stable tarball
		  head/                 # Fossil repo working checkout

  The working checkout directory's name doesn't matter, it just needs
  to be next to the stable tree so that ../4.0.0 finds it.

  This dependence on an existing ABI dump file is deemed reasonable
  since the ABI of the stable version had better not be changing!
  Plus, it saves some processing time, since ACC can load the stable
  ABI info without re-parsing its headers and library file.

- Re-bootstrap the system in pedantic mode, then do a clean rebuild.
  Fix any new errors and warnings.

  Known bogus warnings:

  - Query's std::basic_ios<> base class is not being initialized.

    Yes, we know.  We don't care.

  - The "==" float comparisons in lib/stadapter.cpp are harmless.

    They're comparisons against special NaN and infinity constants.
    Those are safe.

- Re-bootstrap it again without "pedantic", to avoid shipping the
  pedantic build files.
|

|
|
<
<
<

|

|
|
|

|

|

|
|
|
|

|
|

|
|
|
|

|
|

|

<
>
|

<
>
|
|
<
<
<
1
2
3
4



5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

34
35
36

37
38
39



-   Change version number in the top-level `CMakeLists.txt`.

    All other places the version number occurs are files generated
    by CMake from a corresponding `*.in` file.




-   Run `make abicheck`.  There should be no changes.

    You may have to run the following command in the current "stable"
    directory before this will succeed, since it depends on there being
    an ACC dump file in place already.

        $ abi-compliance-checker -lib tabula -dump abi.xml

    This assumes a working tree that looks something like this:

        projects/
           libtabula/
              4.0.0/                # unpacked stable tarball
              head/                 # Fossil repo working checkout

    The working checkout directory's name doesn't matter, it just needs
    to be next to the stable tree so that ../4.0.0 finds it.

    This dependence on an existing ABI dump file is deemed reasonable
    since the ABI of the stable version had better not be changing!
    Plus, it saves some processing time, since ACC can load the stable
    ABI info without re-parsing its headers and library file.

-   Reconfigure the build system in pedantic mode, then do a clean
    rebuild.  Fix any new errors and warnings.

    Known bogus warnings:


    -   `Query`'s `std::basic_ios<>` base class is not being
        initialized.  Yes, we know.  We don't care.


    -   The `==` float comparisons in `lib/stadapter.cpp` are harmless.
        They're comparisons against special `NaN` and `infinity``
        constants.  Those are safe.



Changes to Wishlist.md.