Allow arbitrary configuration options to be set from the command line…

… (fix issue 893) git-svn-id: https://tesseract-ocr.googlecode.com/svn/trunk@837 d0cd1f9f-072b-0410-8dd7-cf729c803f20
LittleDW · Apr 29, 2013 · 7dcfd02 · 7dcfd02
1 parent 1032cb1
commit 7dcfd02
Show file tree

Hide file tree

Showing 2 changed files with 31 additions and 6 deletions.
diff --git a/api/tesseractmain.cpp b/api/tesseractmain.cpp
@@ -90,6 +90,9 @@ int main(int argc, char **argv) {
     } else if (strcmp(argv[arg], "--print-parameters") == 0) {
       noocr = true;
       print_parameters = true;
+    } else if (strcmp(argv[arg], "-o") == 0 && arg + 1 < argc) {
+      // handled properly after api init
+      ++arg;
     } else if (image == NULL) {
       image = argv[arg];
     } else if (output == NULL) {
@@ -105,7 +108,8 @@ int main(int argc, char **argv) {
 
   if (output == NULL && noocr == false) {
     fprintf(stderr, _("Usage:%s imagename outputbase|stdout [-l lang] "
-                      "[-psm pagesegmode] [configfile...]\n\n"), argv[0]);
+                      "[-psm pagesegmode] [-o configvar=value] "
+                      "[configfile...]\n\n"), argv[0]);
     fprintf(stderr,
             _("pagesegmode values are:\n"
               "0 = Orientation and script detection (OSD) only.\n"
@@ -119,8 +123,9 @@ int main(int argc, char **argv) {
               "8 = Treat the image as a single word.\n"
               "9 = Treat the image as a single word in a circle.\n"
               "10 = Treat the image as a single character.\n"));
-    fprintf(stderr, _("-l lang and/or -psm pagesegmode must occur before any"
-                      "configfile.\n\n"));
+    fprintf(stderr, _("multiple -o arguments are allowed.\n"));
+    fprintf(stderr, _("-l lang, -psm pagesegmode and any -o options must occur"
+                      "before any configfile.\n\n"));
     fprintf(stderr, _("Single options:\n"));
     fprintf(stderr, _("  -v --version: version info\n"));
     fprintf(stderr, _("  --list-langs: list available languages for tesseract "
@@ -143,6 +148,21 @@ int main(int argc, char **argv) {
     exit(1);
   }
 
+  char opt1[255], opt2[255];
+  for (arg = 0; arg < argc; arg++) {
+    if (strcmp(argv[arg], "-o") == 0 && arg + 1 < argc) {
+      strncpy(opt1, argv[arg + 1], 255);
+      *(strchr(opt1, '=')) = 0;
+      strncpy(opt2, strchr(argv[arg + 1], '=') + 1, 255);
+      opt2[254] = 0;
+      ++arg;
+
+      if(!api.SetVariable(opt1, opt2)) {
+        fprintf(stderr, _("Could not set option: %s=%s\n"), opt1, opt2);
+      }
+    }
+  }
+
   if (list_langs) {
      GenericVector<STRING> languages;
      api.GetAvailableLanguagesAsVector(&languages);

diff --git a/doc/tesseract.1 b/doc/tesseract.1
@@ -31,7 +31,7 @@
 tesseract \- command\-line OCR engine
 .SH "SYNOPSIS"
 .sp
-\fBtesseract\fR \fIimagename\fR \fIoutbase\fR [\fI\-l lang\fR] [\fI\-psm N\fR] [\fIconfigfile\fR \&...]
+\fBtesseract\fR \fIimagename\fR \fIoutbase\fR|\fIstdout\fR [\fI\-l lang\fR] [\fI\-psm N\fR] [\gI\-o configvar=value\fR] [\fIconfigfile\fR \&...]
 .SH "DESCRIPTION"
 .sp
 tesseract(1) is a commercial quality OCR engine originally developed at HP between 1985 and 1995\&. In 1995, this engine was among the top 3 evaluated by UNLV\&. It was open\-sourced by HP and UNLV in 2005, and has been developed at Google since then\&.
@@ -45,7 +45,7 @@ The name of the input image\&. Most image file formats (anything readable by Lep
 \fIoutbase\fR
 .RS 4
 The basename of the output file (to which the appropriate extension will be appended)\&. By default the output will be named
-\fIoutbase\&.txt\fR\&.
+\fIoutbase\&.txt\fR\&. When stdout is used as outbase, output will be sent to stdout\&.
 .RE
 .PP
 \fI\-l lang\fR
@@ -80,6 +80,11 @@ are:
 .\}
 .RE
 .PP
+\fI\-o configvar=value\fR
+.RS 4
+Sets a configuration variable\&. Multiple options can be set by using -o multiple times, once for each option\&.
+.RE
+.PP
 \fI\-v\fR
 .RS 4
 Returns the current version of the tesseract(1) executable\&.
@@ -102,7 +107,7 @@ hocr \- Output in hOCR format instead of as a text file\&.
 .RE
 .RE
 .sp
-\fBNota Bene:\fR The options \fI\-l lang\fR and \fI\-psm N\fR must occur before any \fIconfigfile\fR\&.
+\fBNota Bene:\fR The options \fI\-l lang\fR, \fI\-psm N\fR and \fI\-o configvar=value\fR must occur before any \fIconfigfile\fR\&.
 .SH "LANGUAGES"
 .sp
 There are currently language packs available for the following languages: