Fixed limitation of length of sequence, updated readlength distribution to jagged array

doc/formats.tex

@@ -1,9 +1,9 @@
 \documentclass[10pt]{article}
-\usepackage{color}
+\usepackage{xcolor}
 \definecolor{gray}{rgb}{0.7,0.7,0.7}
 \usepackage{framed}
 \usepackage{enumitem}
-\usepackage{longtable}
+\usepackage{adjustbox}
 \usepackage[pdfborder={0 0 0},hyperfootnotes=false]{hyperref}
 
 \addtolength{\textwidth}{3.4cm}
@@ -243,12 +243,67 @@ \section{.rlens.gz}
 Readlength distribution. Distribution is count of alignments of
 specific readlengths. Depending on runmode there might be multiple
 groups as taxid/refs and there will be group specific
-distributions. Distributions for different groups are split by
+distributions.
+\subsection{Old versions (pre april 2026)}
+Distributions for different groups are split by
 newlines. First entry on each line is the identifer
 (chromosomename,taxid). The remaining entries are the number of times
 we have observed an alignment of length (columnnumber). Notice that
 the first 30 column of counts is likely to be zero since reads shorter
 than 30basepairs are normally discarded.
+\subsection{Newer versions: human nice format}
+Each distribution is a newline. Each entry in the distribution is bin:count seperated by tab.
+Example can be seen below. This is the default option that should be enforcable by \colorbox{gray!10}{\texttt{--rlens\_flat\_out 0}}. Output below is clipped at rightside and at bottom. Example shown is from one of the testfiles, and has been generated by \colorbox{gray!10}{\texttt{cut -c1-80 test/output/test\_getdamage\_local.rlens|head }}
+\begin{framed}
+  \small
+\begin{verbatim}
+id	rlen:count
+0	35:33	36:29	37:24	38:38	39:24	40:29	41:36	42:28	43:27	44:36	45:29	46:23	47:32...
+1	35:103	36:106	37:102	38:111	39:109	40:126	41:125	42:164	43:154	44:133	45:152...
+2	35:35	36:44	37:28	38:41	39:33	40:36	41:41	42:33	43:34	44:43	45:41	46:36	47:40...
+3	35:36	36:39	37:25	38:37	39:29	40:34	41:39	42:33	43:34	44:43	45:41	46:33	47:39...
+4	35:100	36:98	37:95	38:112	39:107	40:136	41:127	42:161	43:157	44:138	45:134	46:...
+5	35:92	36:118	37:110	38:125	39:129	40:168	41:158	42:164	43:195	44:180	45:177	46...
+6	35:95	36:130	37:111	38:125	39:129	40:178	41:167	42:148	43:196	44:193	45:185	46...
+7	35:138	36:145	37:131	38:125	39:147	40:179	41:190	42:205	43:196	44:201	45:220	4...
+8	35:95	36:132	37:113	38:137	39:133	40:182	41:162	42:176	43:196	44:190	45:189	46...
+...
+\end{verbatim}
+\end{framed}
+\subsection{Newer versions: human nice format}
+
+
+The parsing of the standard output rlens file can be abit confusing for programs to parse due to the varying columnwidth. Therefore by supplying  \colorbox{gray!10}{\texttt{--rlens\_flat\_out 1}} or \colorbox{gray!10}{\texttt{-z 1}} in relevant commands the output will be flattened in to a simpler but longer format.
+\begin{framed}
+  \small
+\begin{verbatim}
+IFILE=./data/f570b1db7c.dedup.filtered.rname.bam
+../metaDMG-cpp getdamage -r 1 -l 35 -p 5 -o output/test_getdamage_local ${IFILE} -z 1
+gunzip -c output/test_getdamage_local.rlens.gz|head -n20
+##below is the output from the last command above
+ID	rlen	count
+0	35	33
+0	36	29
+0	37	24
+0	38	38
+0	39	24
+0	40	29
+0	41	36
+0	42	28
+0	43	27
+0	44	36
+0	45	29
+0	46	23
+0	47	32
+0	48	23
+0	49	25
+0	50	17
+0	51	16
+0	52	19
+0	53	17
+\end{verbatim}
+\end{framed}
+
 \end{document}
 
 %%% Local Variables:

main_dfit.cpp

@@ -25,7 +25,7 @@
 #include "dfit_helppage.h"
 #include "mrand.h"
 
-extern htsFormat *dingding2;
+//extern htsFormat *dingding2;
 
 // the compare function for double values
 static int compare (const void * a, const void * b)
@@ -672,6 +672,7 @@ int main_dfit(int argc, char **argv) {
     double CI = 0.95;
     int doCI = 2;
     int rng_type = -1;
+    htsFormat *dingding2 = (htsFormat *)calloc(1, sizeof(htsFormat));
     while (*(++argv)) {
         if (strcasecmp("-h", *argv) == 0)
           HelpPageSimple(stderr);
@@ -947,7 +948,7 @@ int main_dfit(int argc, char **argv) {
       free(outfile_name);
     if(lib_prep)
       free(lib_prep);
-
+    free(dingding2);
   gettimeofday(&end_time, NULL);
   long seconds = end_time.tv_sec - start_time.tv_sec;
   long microseconds = end_time.tv_usec - start_time.tv_usec;

main_pmd.cpp

@@ -15,8 +15,8 @@
 
 #include "profile.h"  // for reconstructRefWithPosHTS
 
-extern htsFormat *dingding2;
-htsFormat *dingding3 = dingding2;
+//extern htsFormat *dingding2;
+//htsFormat *dingding3 = dingding2;
 
 
 int nproc = 0;  // number of reads processed
@@ -159,7 +159,7 @@ void parse_sequencingdata(char *refName, char *fname, int mapped_only, int se_on
     kstr->s = NULL;
     mypair.first = kstr;
     samFile *in = NULL;
-
+    htsFormat *dingding3 = (htsFormat *)calloc(1, sizeof(htsFormat));
     if (refName) {
         char *ref = (char *)malloc(10 + strlen(refName) + 1);
         snprintf(ref,10 + strlen(refName) + 1, "reference=%s", refName);
@@ -241,6 +241,7 @@ void parse_sequencingdata(char *refName, char *fname, int mapped_only, int se_on
     sam_hdr_destroy(hdr);
     sam_close(in);
     free(fname);
+    free(dingding3);
 }
 
 int usage(FILE *fp, int val) {