Version 1.5.0, Release Candidate 0

libxls · Jan 31, 2019 · 79f9ea9 · 79f9ea9
1 parent 43b6d47
commit 79f9ea9
Show file tree

Hide file tree

Showing 6 changed files with 98 additions and 42 deletions.
diff --git a/Makefile.am b/Makefile.am
@@ -18,6 +18,8 @@ nobase_include_HEADERS = \
 bin_PROGRAMS = xls2csv
 noinst_PROGRAMS = test_libxls test2_libxls
 
+dist_man1_MANS = man/xls2csv.man
+
 lib_LTLIBRARIES = libxlsreader.la
 
 xls2csv_SOURCES = \
@@ -49,6 +51,11 @@ test2_libxls_LDADD = libxlsreader.la
 check_PROGRAMS = test_libxls test2_libxls
 TESTS = test_libxls
 
+EXTRA_DIST = test/files/test2.xls
+
+distclean-local:
+	-rm -f $(builddir)/test.htm
+
 EXTRA_PROGRAMS = fuzz_xls
 
 # Force C++ linking for fuzz target
@@ -58,10 +65,3 @@ fuzz_xls_SOURCES = fuzz/fuzz_xls.c
 fuzz_xls_CFLAGS = -Wall -Werror -pedantic-errors -std=c99
 fuzz_xls_LDADD = libxlsreader.la @LIB_FUZZING_ENGINE@
 fuzz_xls_LDFLAGS = -static
-
-# Extra files to distribute
-EXTRA_DIST = doc test/files/test2.xls
-
-dist-hook:
-	rm -rf `find $(distdir)/doc -name CVS`
-
diff --git a/README.md b/README.md
@@ -5,21 +5,24 @@ libxls - Read XLS files from C
 ==
 
 This is libxls, a C library for reading Excel files in the nasty old binary OLE
-format. **We are in the process of preparing a 1.5 release and moving the project
-over from [SourceForge](https://sourceforge.net/projects/libxls/).** If you need
-a stable release, head back to SourceForge, or see the
-[releases](https://github.com/libxls/libxls/releases) section, which currently
-has copies of everything from SourceForge.
+format, plus a command-line tool for converting XLS to CSV (named, appropriately
+enough, `xls2csv`).
 
-Please note that the current stable releases have several public security
-vulnerabilities. We'll have them fixed in 1.5. Hang tight.
+After several years of neglect, libxls is under new management as of the 1.5.x
+series. Head over to [releases](https://github.com/libxls/libxls/releases) to
+get the latest stable version of libxls 1.5, which fixes *many* security
+vulnerabilities found in libxls 1.4 and earlier.
 
-Changes since 1.4:
+Libxls 1.5 also includes new APIs for parsing files stored in memory buffers,
+and returns errors instead of exiting upon encountering malformed input. If you
+find a bug, please file it on the [GitHub issue tracker](https://github.com/libxls/libxls/issues).
+
+Changes to libxls since 1.4:
 
 * Hosted on GitHub (hooray!)
-* New in-memory parsing API
+* New in-memory parsing API (see `xls_open_buffer`)
 * Internals rewritten to return errors instead of exiting
-* Heavily fuzz-tested with clang's libFuzzer, fixing many memory leaks and *cough* CVEs
+* Heavily fuzz-tested with clang's libFuzzer, fixing many memory leaks and CVEs
 * Improved compatibility with C++
 * Continuous integration tests on Mac, Linux, and Windows
 * Lots of other small fixes, see the commit history
@@ -31,35 +34,44 @@ xls_error_t error = LIBXLS_OK;
 xlsWorkBook *wb = xls_open_file("/path/to/finances.xls", "UTF-8", &error);
 if (wb == NULL) {
     printf("Error reading file: %s\n", xls_getError(error));
-} else {
-    for (int i=0; i<wb->sheets.count; i++) { // sheets
-        xl_WorkSheet *work_sheet = xls_getWorkSheet(work_book, i);
-        error = xls_parseWorkSheet(work_sheet);
-        for (int j=0; j<=work_sheet->rows.lastrow; j++) { // rows
-            xlsRow *row = xls_row(work_sheet, j);
-            for (int k=0; k<=work_sheet->rows.lastcol; k++) { // columns
-                xlsCell *cell = &row->cells.cell[k];
-                // do something with cell
-                if (cell->id == XLS_RECORD_FORMULA) { // formula
-                } else if (cell->l == 0) { // its a number
-                    ... use cell->d
+    exit(1);
+}
+for (int i=0; i<wb->sheets.count; i++) { // sheets
+    xl_WorkSheet *work_sheet = xls_getWorkSheet(work_book, i);
+    error = xls_parseWorkSheet(work_sheet);
+    for (int j=0; j<=work_sheet->rows.lastrow; j++) { // rows
+        xlsRow *row = xls_row(work_sheet, j);
+        for (int k=0; k<=work_sheet->rows.lastcol; k++) { // columns
+            xlsCell *cell = &row->cells.cell[k];
+            // do something with cell
+            if (cell->id == XLS_RECORD_BLANK) {
+                // do something with a blank cell
+            } else if (cell->id == XLS_RECORD_NUMBER) {
+               // use cell->d, a double-precision number
+            } else if (cell->id == XLS_RECORD_FORMULA) {
+                if (strcmp(cell->str, "bool") == 0) {
+                    // its boolean, and test cell->d > 0.0 for true
+                } else if (strcmp(cell->str, "error") == 0) {
+                    // formula is in error
                 } else {
-                    if(cell->str == "bool") // its boolean, and test cell->d > 0.0 for true
-                    if(cell->str == "error") // formula is in error
-                    else ... cell->str is valid as the result of a string formula.
+                    // cell->str is valid as the result of a string formula.
                 }
+            } else if (cell->str != NULL) {
+                // cell->str contains a string value
             }
         }
-        xls_close_WS(work_sheet);
     }
-    xls_close_WB(wb);
+    xls_close_WS(work_sheet);
 }
+xls_close_WB(wb);
 ```
 
 The library also includes a CLI tool for converting Excel files to CSV:
 
     ./xls2csv /path/to/file.xls
 
+The man page for `xls2csv` has more details.
+
 Libxls should run fine on both little-endian and big-endian systems, but if not
 please open an [issue](https://github.com/libxls/libxls/issues/new).
 
@@ -71,7 +83,7 @@ Installation
 If you want a stable version, check out the
 [Releases](https://github.com/libxls/libxls/releases) section, which has copies of everything
 you'll find in [Sourceforge](https://sourceforge.net/projects/libxls/files/),
-and download version 1.4.0.
+and download version 1.5.0 or later.
 
 For full instructions see [INSTALL](INSTALL), or here's the tl;dr:
 

diff --git a/configure.ac b/configure.ac
@@ -1,9 +1,9 @@
-AC_INIT([libxls],[1.4.0],[emmiller@gmail.com])
+AC_INIT([libxls],[1.5.0-rc0],[emmiller@gmail.com])
 AC_CONFIG_SRCDIR([test/test.c])
 PACKAGE=libxls
 
 LIBXLS_MAJOR_VERSION=1
-LIBXLS_MINOR_VERSION=4
+LIBXLS_MINOR_VERSION=5
 LIBXLS_MICRO_VERSION=0
 
 VERSION=$LIBXLS_MAJOR_VERSION.$LIBXLS_MINOR_VERSION.$LIBXLS_MICRO_VERSION
@@ -28,7 +28,7 @@ AC_CANONICAL_BUILD
 AC_CANONICAL_HOST
 
 AC_CONFIG_HEADERS([config.h])
-AC_PREFIX_DEFAULT(/usr/local/libxls)
+AC_PREFIX_DEFAULT(/usr/local)
 
 AC_PROG_INSTALL
 AC_PROG_CC

diff --git a/man/xls2csv.man b/man/xls2csv.man
@@ -0,0 +1,41 @@
+.TH XLS2CSV 1 "31 January 2019"
+.SH NAME
+xls2csv \- convert binary Excel (.xls) files to CSV
+.SH SYNOPSIS
+.B xls2csv
+.IR input-file
+[\fB-l\fR]
+[\fB-v\fR]
+[\fB-e\fR \fIencoding\fR]
+[\fB-t\fR \fIsheet\fR]
+[\fB-q\fR \fIquote char\fR]
+[\fB-f\fR \fIfield separator\fR]
+.P
+.SH DESCRIPTION
+.B xls2csv
+converts binary Excel files (.xls) to comma-separated values (CSV), sending the plain-text output to standard out.
+.PP
+Unless otherwise specified, all sheets in the workbook are processed, and the output values are double-quoted, separated by semicolons, and encoded as UTF-8. This behavior may be customized with the options below.
+.SH OPTIONS
+.TP
+.BR \-l
+List sheets contained in the file but do not output their contents.
+.TP
+\fB\-t\fR \fIsheet\fR
+Only process the named sheet.
+.TP
+\fB-e\fR \fIencoding\fR
+The iconv output encoding (default "UTF-8")
+.TP
+\fB-q\fR \fIquote char\fR
+The character used to quote strings (default '"')
+.TP
+\fB-f\fR \fIfield separator\fR
+The character used to separate strings (default ';')
+.TP
+\fB-v\fR
+Verbose mode
+.SH BUGS
+The conversion tool has not been tested on big-endian machines.
+.SH AUTHOR
+Copyright (C) 2004-2019 the libxls authors
diff --git a/test/test.c b/test/test.c
@@ -38,7 +38,7 @@
 #include <string.h>
 #include <ctype.h>
 
-#include "xls.h"
+#include "../include/xls.h"
 
 int main(int argc, char *argv[])
 {
@@ -51,10 +51,13 @@ int main(int argc, char *argv[])
     xls_error_t code = LIBXLS_OK;
     struct st_row_data* row;
     WORD t,tt;
-    pWB=xls_open_file("test/files/test2.xls", "UTF-8", &code);
+    char *srcdir = getenv("srcdir");
+    char path[2048];
+    snprintf(path, sizeof(path), "%s/test/files/test2.xls", srcdir ? srcdir : ".");
+    pWB=xls_open_file(path, "UTF-8", &code);
 
     if (pWB == NULL) {
-        fprintf(stderr, "Unable to open file: %s\n", xls_getError(code));
+        fprintf(stderr, "Unable to open %s: %s\n", path, xls_getError(code));
         return 1;
     }
 

diff --git a/test/test2.c b/test/test2.c
@@ -40,7 +40,7 @@
 #include <string.h>
 #include <ctype.h>
 
-#include "xls.h"
+#include "../include/xls.h"
 
 int main(int argc, char *argv[])
 {