generated by libzint. Whether or not those barcodes are correct should be
validated in the test suite for libzint, for qZint it's good enough to
produce the same printouts that libzint does, which isn't trivial, especially
as some API names have different meanings, such as qZint ```text```
is libzint ```data```. Props Alonso Schaich <alonso@freebsd.org>
and set Escaped flag; fix tooltip that they're converted to spaces;
sequence dialog: re-organize to put Create Sequence stuff only in groupbox
and put Import -> From File and Clear at bottom, similar to data dialog;
mainwindow: use new Escaped flag from data dialog and set checkbox and
statusbar message accordingly
and restrict sequence to max 10,000, add button icons;
make Export dialog sizable and show every 100 results, add button icon;
fix saving CHANNEL option, fix Export painting main window (Windows), fix
guard descent not-resetting
qzint: add getAsCLI(), warnLevel(), extra isStackable()/isComposite() etc
Add ZBarcode_BarcodeName()
manual: doc above, some fixes, tweaks
The ```configuration``` is required by the visual studio generator to determine
whether to use a Debug or a Release setup (or a user customized one). It's
required by whatever uses that generator rather then just the GUI code.
Add instructions for the Visual Studio Project File generator to the README
and enable the test on Windows.
The ```addon_row_height``` variable is conditionally initialized in some
branches, which also set ```addon_latch``` to one(true). Later in the loop's
body, ```addon_row_height``` is accessed if ```addon_latch``` is true.
Unlike ```addon_row_height```, ```addon_latch``` is defined outside of the
loop's body, and as it is never reset to zero, keeps it's value of one once
it has been assigned. Future iterations of the loop can therefore not branch
into the code that initializes ```addon_row_height``` and (re)assigns
```addon_latch```, and ```addon_latch``` will be true, which causes
```addon_row_height``` to be accessed without having been initialized.
On most platforms, the ```addon_row_height``` variable will always be
allocated on the same memory, while skipping the initialization causes
the value that was previously assigned to ```addon_row_height``` to
remain on that memory.
Pull the variable declaration out of the for loop's body to assure the
previous iteration's value remains there independend of compiler or
platform specific behaviour.
other than for digits (worse for TEX 7-11 embedded in extended) and for
returning X12/EDI at EOD - will cause changes in encodation;
allow for GS1 GS in B256 (should never happen);
some re-jigging of dm200encode and fiddling with dm_isXXX;
remove some DM_DEBUG from dm_placementbit
The `__has_feature` macro is a clang extension that works like a macro. If
zint is being compiled using a compiler other then clang, `__has_feature`
is not defined. As it is not defined, it cannot be used as a function call
expression. Any environment that doesn't have `__has_feature` should prevent
the preprocessor from "seeing" the invokation of the macro, as invoking
undefined macros is not supported by the C language.
The usual procedure to assure this would be a construction like
\#if defined(__has_feature)
\# if __has_feature(...)
\# endif
\#endif
which, combined with the GCC check we have here, would result in a
3 level nesting of if and elseif expressions .. and that's without
covering microsoft's compiler.
For this purpose, Qt >= 5.13 has been defining the `__has_feature`
macro on non-clang environments, while defining all (clang) features
to not be available. This allows to write feature checks with "less"
nesting, as we have here.
Mimic Qt-5.13's behavior and provide the `__has_feature` macro if
it's not provided otherwise (either by clang or by Qt), allowing the
function-call-like expression to be parsed on those systems.
PDF417 & CODABLOCKF due to height/scale rounding by changing
out_large_bar_height() to return scaled int values for raster,
props codemonkey82 (#204)
raster/vector.c: const some vars
vector.c: rect_count, last_start_row UPC/EAN only
library.c: check for stacking symbols >= 200
add specify rows option (option_3) (#204);
warn if cols increased from specified (back-incompatible);
move table definitions from "pdf417.h" to new "pdf417_tabs.h" and
make INTERNAL_DATA and share with composite.c (saves ~10K);
prefix routines and tables with "pdf_";
some small performance improvements through if/elses, pdf_textprocess()
& pdf_numbprocess() loop simplifications
MICROQR: fix debug access crash on printing non-NUL-terminating binary
DATAMATRIX: fix missing ++ from "[tp]" at C40/TEXT EOD processing of GS1
(though probably never reached); use "[tp++]" throughout
Add const to static tables missing it and also to some variables
Change "debug" -> "debug_print" throughout
- use fixed-length string tables (mostly) instead of (char *) pointer ones
(saves ~40K)
- re-use C128Table for CODABLOCKF and CODE16K
(required removal of Stop character and extra CODE16K-only entry)
- use pointer to destination and copy (memcpy/strcpy(), bin_append_posn())
instead of concatenating (strcat()) (mostly)
- replace last remaining bin_append()s with bin_append_posn();
bin_append() removed
- add length arg to toupper() and expand() (avoids strlen())
- change is_sane() to use table-based flags (avoids an iteration)
- rename lookup() to is_sane_lookup() and change to check and return posns
and use in pointer to destination loops (avoids strcat()s)
- remove special case PHARMA in expand() (dealt with in pharma())
- make #define SILVER/CALCIUM/TECHNETIUM/KRSET etc static strings
- replace strchr() -> posn()
- CODE128: populate destination once in checksum loop; re-use and export
some more routines (c128_set_a/b/c(), c128_put_in_set()) for sharing;
prefix defines (SHIFTA -> C128_SHIFTA etc) and existing exported routines
- use factor XOR toggle trick in checksum calcs (avoids branch)
- raster.c: fill out single 1-pixel row and copy using new draw_bar_line(),
copy_bar_line() routines; similarly in buffer_plot compare previous line &
copy if same (same technique as used to improve non-half-integer scaling,
significant performance increase, (c) codemonkey82);
also done for PNG (BMP/GIF/PCX/TIFF not done)
- raster/vector/output.c: shorten "output_" prefix -> "out_";
sync vector to other raster changes to try to keep source files similar
- 2of5.c: prefix "c25_"
JAPANPOST: return error if input data truncated (backward incompatible)
DAFT: max chars 50 -> 100
common.c: istwodigit() -> is_twodigit()
common.c/emf.c/output.c: use some further stripf()s (MSVC6 float variations)
library.c: new check_output_args() helper
zint.h: add BARCODE_LAST marker and use in library.c
QRCODE: remove a NOLINT (requires clang-tidy-13), one remaining
CMake: separate no-optimize from ZINT_DEBUG into new ZINT_NOOPT option
Rename barcode funcs to same as BARCODE_XXX name
library: barcode funcs array for dispatch, used for ZBarcode_ValidID() also
general: change is_sane() comparison to nonzero from ZINT_ERROR_INVALID_OPTION
MAILMARK: fuller error messages
CODABAR: add option to show check character in HRT
zint.h: use 0xNNNN for OR-able defines
GUI: add guard descent height reset button, add Zint version to window title,
static get_zint_version() method, use QStringLiteral (QSL shorthand),
use SIGNAL(toggled()), add errtxt "popup" and status bar, add icons,
add saveAs shortcut, add main menu, context menus and actions, add help,
reset_view() -> reset_colours(), add copy to clipboard as EMF/GIF/PNG/TIF,
lessen triggering of update_preview(), shorten names of getters/setters,
simplify/shorten some update_preview() logic in switch,
CODEONE disable structapp for Version S
qzint.cpp: add on_errored signal, add missing getters, add test
GRIDMATRIX, MAXICODE, MICROPDF417, PDF417, QRCODE, ULTRA
DOTCODE: use pre-calculated generator poly coeffs in Reed-Solomon for
performance improvement
PDF417/MICROPDF417: use common routine pdf417_initial()
GUI: code lines <= 118, shorthand widget_obj(),
shorten calling upcean_addon_gap(), upcean_guard_descent()
various backend: var name debug -> debug_print
also improve non-half-int interpolation performance
raster/vector.c: use new stripf() func in "common.c" to workaround gcc
32-bit float calculation variations
gs1.c: allow dummy AI "[]" if GS1NOCHECK_MODE and has data (#204);
also add note re TPX AI 235 and terminating FNC1
Remove trailing whitespace in various files
also avoid distributive multiplication with floats to lessen chances of
platform variation (#204 ARM-Cortex crash)
- raster.c: Don't allow for text if scale < 1.0
- raster.c: Cast some indexes to (size_t) to allow for large scale
- vector.c: Check malloc()s and return ZINT_ERROR_MEMORY on fail
- raster/vector.c: various var name changes & other code fiddling
- library.c: Check that scale/height/whitespace/border are reasonable values:
scale (0.01-100), height (0-500), whitespace_width/height (0-100),
border_width (0-100)
- CLI: allow both e.g. '-height' and '--height' (getopt_long_only())
- gif.c: fix GIF_ZLW_PAGE_SIZE -> GIF_LZW_PAGE_SIZE
- GUI: allow whitespace/scale to 100
