This commit adds an option to build using CMake by adding
CMakeLists.txt. CMake builds are consistent with the builds currently
produced.

For Linux build, CPack produces artifacts consistent with the current
RPM specification.

CMake is also added to the CI pipeline in main.yml.

* cli/gcdmodel.cpp: Fix typos in comments.

* cli/gcdmodel.cpp (CGcdData::fixModelAndSubmodelOrder): Fix typos in comments.

* cli: Support "max" as an alias to the maximum value for the order switch.

* cli/cmdline.cpp (parseArg): Understand "all" for the order switch.
(showUsage): Document it.
* cli/gcdmodel.cpp (CGcdData::fixModelAndSubmodelOrder): Ensure the
root model order gets recomputed when "/o:max" was provided as argument.
* cli/model.h (UNDEFINED_ORDER): Redefine as the minimum int value.
(MAXIMUM_ORDER): New constant.
* doc/pict.md: Document it.
* test/arg/.tests: Add test.

Co-authored-by: Maxim Cournoyer <maxim.cournoyer@savoirfairelinux.com>

* Use "nullptr" instead of "NULL"

* Cosmetic fix

* bind2nd() and ptr_fun<>() is removed since C++17

bind2nd() and ptr_fun<>() are deprecated in C++11, and removed in C++17.
These functions are not available in LLVM's "libc++".
Use C++11 lambda expressions for them.

* Break long lines

* Add missing header for std::iswspace()

It's done by Valgrind.

* Use "L" notation for wide character

* Remove unneeded cast

PrintMessage() now uses const pointer

* C++11 accepts std::string for std::wifstream

* C++11 accepts std::string for std::wifstream

* Use mbstowcs()/wcstombs() to convert character class

This fix works when file path includes non-ASCII characters.

* Revert "Use mbstowcs()/wcstombs() to convert character class"

This reverts commit 7fd150a1570b8d6de246da352688daaeadcda998.

mbstowcs()/wcstombs() is not so secure, and recommends mbstowcs_s()/wcstombs_s().
But these secure variants are not supported by some C libraries.

* Fixing DataType enum warnings

* Fixing enum warnings in ValidationErrType, TokenType, Relation, TermDataType, SyntaxErrType

* Fix one last warning

* Fix warning about the array delete

* Fix ErrorCode enum

* Fixing unintalized vars

* Fix warnings pertaining to EncodingType.

* More fixes to enum related warnings

* Enable warnings

* Work without "-fpermissive" option

* Cosmetic fix

* Add "const" to fix warnings

These strings are not needs to modify.

* No need to use "#pragma once" in main C++ source file

* Cosmetic fix

Add newline to end

* Return nullptr if error

* delete() not works for "void*"

Cast specific type to use delete()

* Disable VC++ specific pragmas for other C++ compilers

* Fix warning

* Fix warning

* Use pipeline for compiler

* Enable code optimization

* Add "default:" to fix "switch" syntax warning

* Use "nullptr" for null pointer

* Use nullptr instead of NULL

* Use nullptr instead of NULL

* Use "L" notation for wide character

* Use C++ header instead of raw C header

Multi-byte I/O dose too match for text input in some locale.

For example, "," in numbers treat as decimal separator, not word separator.
This makes "1,234" as "1234", not "1" and "234".

Use "C.UTF-8" to avoid this problem.

* Use UTF-8 for multi-byte character I/O everywhare

This fix is required for "libstdc++".
Without this fix, "libstdc++" only works for 7bit ASCII.

It fixes https://github.com/microsoft/pict/issues/24 .

* Use current locale instead of fixed UTF-8 locale

It is good for non-UTF-8 (ANSI/OEM) encoding input.
Use environment variable "LANG" or "LC_CTYPE" to change input encoding.

(example)
$ LC_CTYPE="C.UTF-8" pict utf8-input.txt

* Add document for non-ASCII (UTF-8/ANSI) input encoding

Signed-off-by: Cleber Rosa <crosa@redhat.com>

Remove openFile entirely to avoid latent problems with stream copy semantics.

Several classes and functions were not const-correct. Did the minimal changes
to fix compilation, although there are other const fixes that can be done.

- replaced for_each with range-based for loop
- added missing std:: scope
- for-loop temp now declared outside the loop to be in scope
- removed casts in calls to trim()