Support for configuration file

docs/ccextractor.cnf.sample

@@ -0,0 +1,369 @@
+# The Input Source tag option give ability to user
+# to take imput from file, standurd input or network
+# This tag take number in its input and there meanings
+# are following
+# 0 = file
+# 1 = stdin
+# 2 = network
+
+INPUT_SOURCE=1
+
+# The Buffer Input tag
+# This tag take number in its input and there meanings
+# are following
+
+BUFFER_INPUT=0
+
+# The Direct Rollup tag
+# This tag take number in its input and there meanings
+# are following
+
+DIRECT_ROLLUP=
+
+#The No font Color Tag
+# This tag take number in its input and there meanings
+# are following
+# 0 = no
+# 1 = yes
+
+NOFONT_COLOR=
+
+#The No type Setting Tag
+# This tag take number in its input and there meanings
+# are following
+# 0 = no
+# 1 = yes
+
+NOTYPE_SETTING=
+
+# The Codec Tag take the preference of codec
+# tag CCX_CODEC_ANY is by default
+# This tag take number in its input and there meaning
+# are following
+# 0 = CCX_CODEC_ANY
+# 1 = CCX_CODEC_TELETEXT
+# 2 = CCX_CODEC_DVB
+
+CODEC=
+
+# The NO Codec Tag do not use codec specified
+# tag CCX_CODEC_NONE is by default
+# This tag take number in its input and there meaning
+# are following
+# 1 = CCX_CODEC_TELETEXT
+# 2 = CCX_CODEC_DVB
+# 3 = CCX_CODEC_NONE
+
+NOCODEC=
+
+# OUTPUT_FORMAT tag specify format of output
+# by default output format is srt
+# This tag take number in its input and there meaning
+# are following
+# 0 = CCX_OF_RAW
+# 1 = CCX_OF_SRT (default)
+# 2 = CCX_OF_SAMI
+# 3 = CCX_OF_TRANSCRIPT
+# 4 = CCX_OF_RCWT
+# 5 = CCX_OF_NULL
+# 6 = CCX_OF_SMPTETT
+# 7 = CCX_OF_SPUPNG
+# 8 = CCX_OF_DVDRAW
+
+OUTPUT_FORMAT=
+
+# Start credit Text is shown at start
+# this tag take text in input
+# Please note text is limited to one line in configuration
+# file
+
+START_CREDIT_TEXT=
+
+# Start credit do not start before apecified time in tag
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+START_CREDIT_NOT_BEFORE=
+
+# Start credits do not start after specified time in tag
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+START_CREDIT_NOT_AFTER=
+
+# Start credits run for at least time specied in tag
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+START_CREDIT_FOR_ATLEAST=
+
+# Start credits are displayed for atmost time specified in tag
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+START_CREDIT_FOR_ATMOST=
+
+# End Credit text are shown at beginning
+# this tag take text in input
+# Please note text is limited to one line in configuration
+# file
+
+END_CREDITS_TEXT=
+
+# End credits run for at least time specied in tag
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+END_CREDITS_FOR_ATLEAST
+
+# End credits are displayed for atmost time specified in tag
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+END_CREDITS_FOR_ATMOST
+
+# Is Video edited or splitted by tool
+# By default its 1, ccextractor will process input files in
+# sequence as if they were all one large file i.e
+# split by a generic, non video-aware tool. If you
+# are processing video hat was split with a editing
+# tool, use 0 so ccextractor doesn't try to rebuild
+# the original timing.
+# This tag take 0 or 1 in input
+
+VIDEO_EDITED=1
+
+# Use GOP for timing instead of PTS. This only applies
+# to Program or Transport Streams with MPEG2 data and
+# overrides the default PTS timing,GOP timing is always
+# used for Elementary Streams.
+#
+# This tag take number in its input and there meaning
+# are following
+#  0 = use pts for time (when reasonable)
+#  1 = use gop for time (when reasonable)
+# -1 = Never use GOP timing (use PTS), even if ccextractor
+#      detects GOP timing is the reasonable choice.
+
+GOP_TIME=0
+
+# Fix padding - some cards (or providers, or whatever)
+# seem to send 0000 as CC padding instead of 8080. If you
+# get bad timing, this might solve it.
+#
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+FIX_PADDINDG=0
+
+# If you hate the repeated lines caused by the roll-up
+# emulation, you can have ccextractor write only one
+# line at a time, getting rid of these repeated lines.
+#
+# This tag take number in its input and there meaning
+# are following
+# 0 = no (default)
+# 1 = yes
+
+NO_ROLL_UP=0
+
+# Roll-up captions can consist of 2, 3 or 4 visible
+# lines at any time (the number of lines is part of
+# the transmission). If having 3 or 4 lines annoys
+# you can use this tag to force the decoder to always
+# use 1, 2 or 3 lines. Note that 1 line is not
+# a real mode rollup mode, so CCExtractor does what
+# it can.
+# In value = 1 the start timestamp is actually the timestamp
+# of the first character received which is possibly more
+# accurate.
+#
+# This tag take number in its input and there meaning
+# are following
+# 1
+# 2
+# 3
+
+FORCED_RU
+
+# This tag trim line
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+TRIM=0
+
+#
+# Report progress and interesting events to stderr
+# in a easy to parse format. This is intended to be
+# used by other programs. See docs directory for details
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+GUI_MODE_REPORTS=0
+
+# Suppress the output of the progress bar
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+NO_PROGRESS_BAR=0
+
+# Sentence capitalization. Use if you hate
+# ALL CAPS in subtitles
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+SENTENCE_CAP=0
+
+#this tag take Dictionary file name containing words in each line
+
+#CAP_FILE=Dictionary.txt
+
+# Program number(pid) to be selected from ts file
+# This tag take number in its input and there meaning
+# are following
+# -1 = automaticaly selected
+#  X = actual program pid in ts file
+
+PROGRAM_NUMBER=-1
+
+# select automaticaly program from ts file
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+AUTO_PROGRAM=0
+
+# Stream number(pid) to be selected from file
+# This tag take number in its input and there meaning
+# are following
+# -1 = automaticaly selected
+#  X = actual stream pid in ts file
+
+STREAM=-1
+
+#Write number of  screenfuls and terminate processing
+#This tag take number of secreen
+
+#SCREEN_TO_PROCESS=100
+
+# start extraction from the time specified
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+START_AT
+
+# End extraction at the time specified
+# this tag only accepts SS, MM:SS or HH:MM:SS
+
+END_AT
+
+# If no CC packets are detected based on the PMT, try
+# to find data in all packets by scanning.
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+INVASTIGATE_PACKET=0
+
+# If the video was recorder using a Hauppauge card, it
+# might need special processing. This parameter will
+# force the special treatment.
+
+HAUPPAUGE_MODE=0
+
+# MP4 files the closed caption data can be embedded in
+# the video track or in a dedicated CC track. If a
+# dedicated track is detected it will be processed instead
+# of the video track. If you need to force the video track
+# to be processed instead use this option.
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+MP4_VIDEO_TRACK
+
+# File need to be encoding in which format
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 0 = CX_ENC_UNICODE
+# 1 = CCX_ENC_LATIN_1
+# 2 = CCX_ENC_UTF_8
+
+ENCODING=2
+
+# Use the pic_order_cnt_lsb in AVC/H.264 data streams
+# to order the CC information.  The default way is to
+# use the PTS information.  Use this switch only when
+# needed.
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+USE_PIC_ORDER=0
+
+# The MythTV branch is needed for analog captures where
+# the closed caption data is stored in the VBI, such as
+# those with bttv cards (Hauppage 250 for example). This
+# is detected automatically so you don't need to worry
+# about this unless autodetection doesn't work for you.
+# This tag take number in its input and there meaning
+# are following
+# 0 = NO Myth
+# 1 = Myth
+# 2 = Auto
+
+AUTO_MYTH=2
+
+# Read the captions from the MPEG2 video stream rather
+# than the captions stream in WTV files
+# This tag take number in its input and there meaning
+# are following
+# 0 = no
+# 1 = yes
+
+WTV_MPEG2=0
+
+# This tag take output file name
+
+#OUTPUT_FILENAME=output.srt
+
+# Don't try to find out the stream for subtitles
+# use the provided pid
+# Takes number as input
+
+#DATA_PID=0x100
+
+# Instead of selecting the stream by its PID, select it
+# by its type (pick the stream that has this type in
+# the PMT
+# Takes number as input
+
+#STREAM_TYPE=0x6
+
+# Assume the data is of this type, don't autodetect. This
+# parameter may be needed if DATA_PID or STREAM_TYPE tag
+# is set and CCExtractor cannot determine how to process
+# the stream. The value may be 2 (MPEG video) or
+# 6 (MPEG private data).
+#
+# Tekes number in input
+
+#TS_FORCED_STREAM_TYPE=0x6
+
+# This tag specify the format of date
+# 0 = ODF_NONE
+# 1 = ODF_HHMMSS
+# 2 = ODF_SECONDS
+# 3 = ODF_DATE
+# 4 = ODF_HHMMSSMS ( HH:MM:SS,MILIS .srt style )
+
+
+#DATE_FORMAT=4

src/ccextractor.c

@@ -4,6 +4,7 @@ License: GPL 2.0
 */
 #include <stdio.h>
 #include "ccextractor.h"
+#include "configuration.h"
 #include <sys/types.h>
 #include <sys/stat.h>
 
@@ -277,6 +278,7 @@ int main(int argc, char *argv[])
 	int show_myth_banner = 0;
 	
 	memset (&cea708services[0],0,63*sizeof (int));
+	parse_configuration(&ccx_options);
 	parse_parameters (argc,argv);
 
 	if (num_input_files==0 && ccx_options.input_source==CCX_DS_FILE)

src/ccextractor.h

@@ -233,9 +233,11 @@ struct ccx_s_teletext_config {
 
 extern LLONG buffered_read_opt (unsigned char *buffer, unsigned int bytes);
 
-//params.cpp
+//params.c
 void parse_parameters (int argc, char *argv[]);
 void usage (void);
+int atoi_hex (char *s);
+int stringztoms (const char *s, struct ccx_boundary_time *bt);
 
 // general_loop.cpp
 void position_sanity_check ();

src/configuration.c

@@ -0,0 +1,148 @@
+#include "ccextractor.h"
+#include "configuration.h"
+#include <stddef.h>
+#define CNF_FILE "ccextractor.cnf"
+struct conf_map
+{
+	char *name;
+	int offset;
+	int (*parse_val)(void *var,char*val);
+};
+
+
+static int set_string(void *var,char*val)
+{
+	char **p = (char**)var;
+	char *val1 = NULL;
+
+	if(val == NULL)
+		return -1;
+
+	val1 = strdup(val);
+	*p = val1;
+	return 0;
+}
+static int set_time(void *var, char*val)
+{
+	struct ccx_boundary_time **p = (struct ccx_boundary_time **)var;
+	if(val == NULL)
+		return -1;
+
+	return stringztoms(val,*p);
+
+}
+static int set_int(void *var, char*val)
+{
+	int *p = (int*)var;
+	if(val == NULL)
+		return -1;
+
+	*p = atoi_hex(val);
+	return 0;
+}
+
+struct conf_map configuration_map[] = {
+	{"INPUT_SOURCE",offsetof(struct ccx_s_options,input_source),set_int},
+	{"BUFFER_INPUT",offsetof(struct ccx_s_options,buffer_input),set_int},
+	{"DIRECT_ROLLUP",offsetof(struct ccx_s_options,direct_rollup),set_int},
+	{"NOFONT_COLOR",offsetof(struct ccx_s_options,nofontcolor),set_int},
+	{"NOTYPE_SETTING",offsetof(struct ccx_s_options,notypesetting),set_int},
+	{"CODEC",offsetof(struct ccx_s_options,codec),set_int},
+	{"NOCODEC",offsetof(struct ccx_s_options,nocodec),set_int},
+	{"OUTPUT_FORMAT",offsetof(struct ccx_s_options,write_format),set_int},
+	{"START_CREDIT_TEXT",offsetof(struct ccx_s_options,start_credits_text),set_string},
+	{"START_CREDIT_NOT_BEFORE",offsetof(struct ccx_s_options,startcreditsnotbefore),set_time},
+	{"START_CREDIT_NOT_AFTER",offsetof(struct ccx_s_options,startcreditsnotafter),set_time},
+	{"START_CREDIT_FOR_ATLEAST",offsetof(struct ccx_s_options,startcreditsforatleast),set_time},
+	{"START_CREDIT_FOR_ATMOST",offsetof(struct ccx_s_options,startcreditsforatmost),set_time},
+	{"END_CREDITS_TEXT",offsetof(struct ccx_s_options,end_credits_text),set_string},
+	{"END_CREDITS_FOR_ATLEAST",offsetof(struct ccx_s_options,endcreditsforatleast),set_time},
+	{"END_CREDITS_FOR_ATMOST",offsetof(struct ccx_s_options,endcreditsforatmost),set_time},
+	{"VIDEO_EDITED",offsetof(struct ccx_s_options,binary_concat),set_int},
+	{"GOP_TIME",offsetof(struct ccx_s_options,use_gop_as_pts),set_int},
+	{"FIX_PADDINDG",offsetof(struct ccx_s_options,fix_padding),set_int},
+	{"NO_ROLL_UP",offsetof(struct ccx_s_options,norollup),set_int},
+	{"FORCED_RU",offsetof(struct ccx_s_options,forced_ru),set_int},
+	{"TRIM",offsetof(struct ccx_s_options,trim_subs),set_int},
+	{"GUI_MODE_REPORTS",offsetof(struct ccx_s_options,gui_mode_reports),set_int},
+	{"NO_PROGRESS_BAR",offsetof(struct ccx_s_options,no_progress_bar),set_int},
+	{"SENTENCE_CAP",offsetof(struct ccx_s_options,sentence_cap),set_int},
+	{"CAP_FILE",offsetof(struct ccx_s_options,sentence_cap_file),set_string},
+	{"PROGRAM_NUMBER",offsetof(struct ccx_s_options,ts_forced_program),set_int},
+	{"AUTO_PROGRAM",offsetof(struct ccx_s_options,ts_autoprogram),set_int},
+	{"STREAM",offsetof(struct ccx_s_options,live_stream),set_int},
+	{"SCREEN_TO_PROCESS",offsetof(struct ccx_s_options,screens_to_process),set_int},
+	{"START_AT",offsetof(struct ccx_s_options,extraction_start),set_time},
+	{"END_AT",offsetof(struct ccx_s_options,extraction_end),set_time},
+	{"INVASTIGATE_PACKET",offsetof(struct ccx_s_options,investigate_packets),set_int},
+	{"FULL_BIN",offsetof(struct ccx_s_options,fullbin),set_int},
+	{"NO_SYNC",offsetof(struct ccx_s_options,nosync),set_int},
+	{"HAUPPAUGE_MODE",offsetof(struct ccx_s_options,hauppauge_mode),set_int},
+	{"MP4_VIDEO_TRACK",offsetof(struct ccx_s_options,mp4vidtrack),set_int},
+	{"ENCODING",offsetof(struct ccx_s_options,encoding),set_int},
+	{"USE_PIC_ORDER",offsetof(struct ccx_s_options,usepicorder),set_int},
+	{"AUTO_MYTH",offsetof(struct ccx_s_options,auto_myth),set_int},
+	{"WTV_MPEG2",offsetof(struct ccx_s_options,wtvmpeg2),set_int},
+	{"OUTPUT_FILENAME",offsetof(struct ccx_s_options,output_filename),set_string},
+	{"OUT_ELEMENTRY_STREAM_FILENAME",offsetof(struct ccx_s_options,out_elementarystream_filename),set_string},
+	{"DATA_PID",offsetof(struct ccx_s_options,ts_cappid),set_int},
+	{"STREAM_TYPE",offsetof(struct ccx_s_options,ts_datastreamtype),set_int},
+	{"TS_FORCED_STREAM_TYPE",offsetof(struct ccx_s_options,ts_forced_streamtype),set_int},
+	{"DATE_FORMAT",offsetof(struct ccx_s_options,date_format),set_int},
+	{NULL}
+};
+static int parse_opts(char *str, struct ccx_s_options *opt)
+{
+	struct conf_map *lmap = configuration_map;
+	char *var =  strtok(str,"=");
+	if(!var)
+		return -1;
+	while(lmap->name)
+	{
+		if(!strcmp(lmap->name,var))
+		{
+			char *val = strtok(NULL,"=");
+			return lmap->parse_val((void*)((char*)opt + lmap->offset ),val);
+		}
+		lmap++;
+	}
+	return 0;
+}
+static void parse_file(FILE *f,struct ccx_s_options *opt)
+{
+	char *str = (char*)malloc(128);
+	char c = '\0';
+	int comments = 0;
+	int i = 0;
+	int ret = 0;
+	while ((c = fgetc(f)) != EOF )
+	{
+		if( c == '\n')
+		{
+			if( str[0] != '\0')
+				ret = parse_opts(str,opt);
+				if(ret < 0)
+					mprint("invalid configuration file\n");
+			comments = 0;
+			i = 0;
+			continue;
+		}
+		if(comments || c == '#')
+		{
+			comments = 1;
+			continue;
+		}
+		str[i] = c;
+		i++;
+	}
+	free(str);
+}
+void parse_configuration(struct ccx_s_options *opt)
+{
+   FILE *f = NULL;
+   if( (f = fopen(CNF_FILE,"r") ) != NULL)
+   {
+		parse_file(f,opt);
+   }
+
+}

src/configuration.h

@@ -0,0 +1,4 @@
+#ifndef _CONFIGURATION_H
+#define _CONFIGURATION_H
+void parse_configuration(struct ccx_s_options *opt);
+#endif

src/params.c

@@ -789,6 +789,28 @@ int atoi_hex (char *s)
 		return atoi (s);
 	}
 }
+void init_option (struct ccx_s_options *option)
+{
+	if(option->gui_mode_reports)
+	{
+		option->no_progress_bar=1;
+		// Do it as soon as possible, because it something fails we might not have a chance
+		activity_report_version();
+	}
+
+	if(option->sentence_cap)
+	{
+		if(add_built_in_words())
+			fatal (EXIT_NOT_ENOUGH_MEMORY, "Not enough memory for word list");
+        if(option->sentence_cap_file && process_cap_file (option->sentence_cap_file))
+            fatal (EXIT_ERROR_IN_CAPITALIZATION_FILE, "There was an error processing the capitalization file.\n");
+
+		shell_sort(spell_lower,spell_words,sizeof(*spell_lower),string_cmp2,NULL);
+		shell_sort(spell_correct,spell_words,sizeof(*spell_correct),string_cmp2,NULL);
+	}
+	if(option->ts_forced_program != -1)
+		option->ts_forced_program_selected = 1;
+}
 
 void parse_parameters (int argc, char *argv[])
 {
@@ -1083,9 +1105,6 @@ void parse_parameters (int argc, char *argv[])
         if (strcmp (argv[i],"--gui_mode_reports")==0)
         {
             ccx_options.gui_mode_reports=1;
-            ccx_options.no_progress_bar=1;
-            // Do it as soon as possible, because it something fails we might not have a chance
-            activity_report_version(); 
 			continue;
         }
         if (strcmp (argv[i],"--no_progress_bar")==0)
@@ -1096,26 +1115,16 @@ void parse_parameters (int argc, char *argv[])
         if (strcmp (argv[i],"--sentencecap")==0 ||
             strcmp (argv[i],"-sc")==0)
         {
-            if (add_built_in_words())
-                fatal (EXIT_NOT_ENOUGH_MEMORY, "Not enough memory for word list");
             ccx_options.sentence_cap=1;
-			shell_sort(spell_lower,spell_words,sizeof(*spell_lower),string_cmp2,NULL);
-			shell_sort(spell_correct,spell_words,sizeof(*spell_correct),string_cmp2,NULL);
 			continue;
         }
         if ((strcmp (argv[i],"--capfile")==0 ||
             strcmp (argv[i],"-caf")==0)
             && i<argc-1)
         {
-            if (add_built_in_words())
-                fatal (EXIT_NOT_ENOUGH_MEMORY, "Not enough memory for word list");
-            if (process_cap_file (argv[i+1])!=0)
-                fatal (EXIT_ERROR_IN_CAPITALIZATION_FILE, "There was an error processing the capitalization file.\n");
             ccx_options.sentence_cap=1;
             ccx_options.sentence_cap_file=argv[i+1];
             i++;
-			shell_sort(spell_lower,spell_words,sizeof(*spell_lower),string_cmp2,NULL);
-			shell_sort(spell_correct,spell_words,sizeof(*spell_correct),string_cmp2,NULL);
 			continue;
         }
         if (strcmp (argv[i],"--program-number")==0 ||