Updates man pages and man->tex conversion; adds rmanprocess.py

configure.ac

@@ -71,7 +71,7 @@ m4_include([m4/ax_pthread.m4])
 ################################################################
 # 2. AC_INIT
 ################################################################
-AC_INIT(Easel, 0.44rc1, [email protected], easel)
+AC_INIT(Easel, 0.44, [email protected], easel)
 AC_MSG_NOTICE([Configuring the Easel library for your system.])
 
 # remember if the user is overriding CFLAGS
@@ -108,7 +108,7 @@ fi
 #     EASEL_VERSION    e.g. "1.9a"
 #
 
-EASEL_DATE="March 2018"
+EASEL_DATE="April 2018"
 EASEL_COPYRIGHT="Copyright (C) 2018 Howard Hughes Medical Institute"
 EASEL_LICENSE="Freely distributed under the BSD open source license."
 EASEL_VERSION=$PACKAGE_VERSION

devkit/rmanprocess.py

@@ -0,0 +1,86 @@
+#! /usr/bin/env python
+
+# rmanprocess.py
+# Massages output of PolyGlotMan's `rman -f latex2e` to fit Tufteian userguide style.
+# Example:
+#    rman -f latex2e hmmbuild.man | rmanprocess.py > manpage.tex
+#
+
+
+import sys
+import re
+
+in_synopsis = False
+
+if len(sys.argv) == 1:
+    f = sys.stdin
+else:
+    f = open(sys.argv[1])
+
+for line in f:
+    line = line.rstrip('\n')
+
+    # State flags (where are we in the document)
+    if re.match(r'\\section{Synopsis}', line):
+        in_synopsis = True
+    elif re.match(r'\\section{', line):
+        in_synopsis = False
+
+    #
+    # Linewise substitutions: replace certain entire lines with something else.
+    #
+    # Remove \documentclass, and changes to \parindent and \parskip
+    if re.match(r'\\documentclass',          line): continue
+    if re.match(r'\\setlength{\\parindent}', line): continue
+    if re.match(r'\\setlength{\\parskip}',   line): continue
+    if re.match(r'\\begin\{document\}',      line): continue
+
+    # Replace \section{Name} with \section{progname - description}, using next line too.
+    if re.match(r'\\section\{Name\}', line):
+        for line in f:
+            if not re.fullmatch(r'\s*', line):
+                break
+        m = re.match(r'(\S+)\s*\\?-\s*(.+)$', line)
+        if m:
+            print(r'\section{{\monob{{{0}}} - {1}}}'.format(m.group(1), m.group(2)))
+        else:
+            print("Error: no progname/description line found");
+            sys.exit(1)
+        continue
+
+    # Remove everything after \section{See Also), and finish.
+    if re.match(r'\\section\{See', line) or re.match(r'\\end\{document', line):
+        print("\\newpage");
+        break
+
+
+    #
+    # Extra directives: preface certain lines with something extra, but still
+    #                   process the line.
+    #
+
+    # In synopsis, put \noindent in front of each commandline, and preserve the .B's as bold.
+    if in_synopsis and re.match(r'\s*\\textbf{', line):
+        line = re.sub(r'\\textbf\{', r'\\monob{', line)
+        print("\\noindent")
+
+    #
+    # Substitutions within a line.
+    # The order of these replacements is important. (More specific ones first.)
+    #
+    line = re.sub(r'\\begin\{itemize\}',  r'\\begin{wideitem}', line)
+    line = re.sub(r'\\end\{itemize\}',    r'\\end{wideitem}',   line)
+    line = re.sub(r'\\section\{',         r'\\subsection*{',    line)   # \subsection* suppresses inclusion in TOC
+    line = re.sub(r'--',                  r'{-}{-}',            line)
+    line = re.sub(r'\\item\s*\[\\textbf', r'\\item [\\monob',   line)   # option names in .TP are emphasized bold
+    line = re.sub(r'\\textbf\{\\% ',      r'\\user{\\% ',       line)   # example command lines are bold, on their own line
+    line = re.sub(r'\\textit\{',          r'\\monoi{',          line)   # metavariables (options, args) are .I in man, mono italic in tex
+    line = re.sub(r'\\textbf\{',          r'\\mono{',           line)   # literals (commands, etc) are .B in man, normal mono in tex
+
+    print(line)
+
+
+
+if f != sys.stdin:
+    f.close()
+

esl_msafile.c

@@ -654,7 +654,10 @@ esl_msafile_GuessFileFormat(ESL_BUFFER *bf, int *ret_fmtcode, ESL_MSAFILE_FMTDAT
   else // if we haven't guessed so far, try selex.
     {				/* selex parser can handle psiblast too */
       if      (fmt_bysuffix == eslMSAFILE_SELEX) *ret_fmtcode = eslMSAFILE_SELEX;
-      else if (msafile_check_selex(bf) == eslOK) *ret_fmtcode = eslMSAFILE_SELEX;
+      else if (msafile_check_selex(bf) == eslOK) {
+	if (fmt_bysuffix == eslMSAFILE_PSIBLAST) *ret_fmtcode = eslMSAFILE_PSIBLAST;
+	else                                     *ret_fmtcode = eslMSAFILE_SELEX;
+      }
       else    ESL_XFAIL(eslENOFORMAT, errbuf, "couldn't guess alignment input format - doesn't even look like selex");
     }
 
@@ -717,13 +720,13 @@ esl_msafile_EncodeFormat(char *fmtstring)
   if (strcasecmp(fmtstring, "stockholm")   == 0) return eslMSAFILE_STOCKHOLM;
   if (strcasecmp(fmtstring, "pfam")        == 0) return eslMSAFILE_PFAM;
   if (strcasecmp(fmtstring, "a2m")         == 0) return eslMSAFILE_A2M;
-  if (strcasecmp(fmtstring, "phylip")      == 0) return eslMSAFILE_PHYLIP;
-  if (strcasecmp(fmtstring, "phylips")     == 0) return eslMSAFILE_PHYLIPS;
   if (strcasecmp(fmtstring, "psiblast")    == 0) return eslMSAFILE_PSIBLAST;
   if (strcasecmp(fmtstring, "selex")       == 0) return eslMSAFILE_SELEX;
   if (strcasecmp(fmtstring, "afa")         == 0) return eslMSAFILE_AFA;
   if (strcasecmp(fmtstring, "clustal")     == 0) return eslMSAFILE_CLUSTAL;
   if (strcasecmp(fmtstring, "clustallike") == 0) return eslMSAFILE_CLUSTALLIKE;
+  if (strcasecmp(fmtstring, "phylip")      == 0) return eslMSAFILE_PHYLIP;
+  if (strcasecmp(fmtstring, "phylips")     == 0) return eslMSAFILE_PHYLIPS;
   return eslMSAFILE_UNKNOWN;
 }
 

esl_sqio.c

@@ -709,16 +709,15 @@ esl_sqio_IsAlignment(int fmt)
 int
 esl_sqio_EncodeFormat(char *fmtstring)
 {
-  if (strcasecmp(fmtstring, "daemon")    == 0) return eslSQFILE_DAEMON;
-  if (strcasecmp(fmtstring, "ddbj")      == 0) return eslSQFILE_DDBJ;
-  if (strcasecmp(fmtstring, "embl")      == 0) return eslSQFILE_EMBL;
   if (strcasecmp(fmtstring, "fasta")     == 0) return eslSQFILE_FASTA;
-  if (strcasecmp(fmtstring, "fmindex")   == 0) return eslSQFILE_FMINDEX;
+  if (strcasecmp(fmtstring, "embl")      == 0) return eslSQFILE_EMBL;
   if (strcasecmp(fmtstring, "genbank")   == 0) return eslSQFILE_GENBANK;
-  if (strcasecmp(fmtstring, "hmmpgmd")   == 0) return eslSQFILE_HMMPGMD;
-  if (strcasecmp(fmtstring, "ncbi")      == 0) return eslSQFILE_NCBI;
+  if (strcasecmp(fmtstring, "ddbj")      == 0) return eslSQFILE_DDBJ;
   if (strcasecmp(fmtstring, "uniprot")   == 0) return eslSQFILE_UNIPROT;
-
+  if (strcasecmp(fmtstring, "ncbi")      == 0) return eslSQFILE_NCBI;
+  if (strcasecmp(fmtstring, "daemon")    == 0) return eslSQFILE_DAEMON;
+  if (strcasecmp(fmtstring, "hmmpgmd")   == 0) return eslSQFILE_HMMPGMD;
+  if (strcasecmp(fmtstring, "fmindex")   == 0) return eslSQFILE_FMINDEX;
   return esl_msafile_EncodeFormat(fmtstring);
 }
 
@@ -741,9 +740,10 @@ esl_sqio_DecodeFormat(int fmt)
   case eslSQFILE_GENBANK:    return "GenBank";
   case eslSQFILE_DDBJ:       return "DDBJ";
   case eslSQFILE_UNIPROT:    return "UniProt";
+  case eslSQFILE_NCBI:       return "NCBI";
   case eslSQFILE_DAEMON:     return "daemon";
   case eslSQFILE_HMMPGMD:    return "hmmpgmd";
-  case eslSQFILE_NCBI:       return "NCBI";
+  case eslSQFILE_FMINDEX:    return "fmindex";
   default:                   break;
   }
   esl_exception(eslEINVAL, FALSE, __FILE__, __LINE__,  "no such sqio format code %d", fmt);

miniapps/esl-afetch.man.in

@@ -1,34 +1,23 @@
-.TH "esl-afetch" 1 "@EASEL_DATE@" "Easel @EASEL_VERSION@" "Easel Manual"
+.TH "esl\-afetch" 1 "@EASEL_DATE@" "Easel @EASEL_VERSION@" "Easel Manual"
 
 .SH NAME
-.TP
-esl-afetch - retrieve alignments from a multi-MSA database
+esl\-afetch \- retrieve alignments from a multi-MSA database
 
 .SH SYNOPSIS
 
-.TP
-Single MSA retrieval:
-.B esl-afetch
-.I [options]
-.I msafile
-.I key
-
-.TP
-Multiple MSA retrieval:
-.B esl-afetch -f
-.I [options]
-.I msafile
-.I keyfile
+.nf
+\fBesl\-afetch\fR [\fIoptions\fR] \fImsafile key\fR
+  (single MSA retrieval)
 
-.TP
-Indexing an MSA file for retrieval:
-.B esl-afetch --index
-.I msafile
+\fBesl\-afetch \-f\fR [\fIoptions\fR] \fImsafile keyfile\fR
+  (multiple MSA retrieval, from a file of keys)
 
+\fBesl\-afetch \-\-index \fImsafile\fR
+  (index an MSA file for retrieval)
 
 .SH DESCRIPTION
 
-.B esl-afetch
+.B esl\-afetch
 retrieves the alignment named
 .I key
 from an alignment database in file
@@ -44,71 +33,75 @@ number (AC).
 
 .PP
 Alternatively, 
-.B esl-afetch -f
+.B esl\-afetch \-f
 provides the ability to fetch many alignments at once.
 The 
-.I -f 
-option has it interpret the second argument as a <keyfile>, a
-file consisting of one name or accession per line.
+.B \-f 
+option has it interpret the second argument as a
+.IR keyfile ,
+a file consisting of one name or accession per line.
 
 .PP
 The
 .I msafile
 should first be SSI indexed with
-.B esl-afetch --index
+.B esl\-afetch \-\-index
 for efficient retrieval. An SSI index is
 not required, but without one alignment retrieval may
 be painfully slow.
 
 .SH OPTIONS
 
 .TP
-.B -h
+.B \-h
 Print brief help; includes version number and summary of
 all options, including expert options.
 
 .TP
-.B -f
+.B \-f
 Interpret the second argument as a 
 .I keyfile
 instead of as just one
-.I key. 
+.IR key . 
 The
 .I keyfile
 contains one name or accession per line.
 This option doesn't work with the
-.B --index
+.B \-\-index
 option.
 
 
 .TP
-.BI -o " <f>"
+.BI \-o " <f>"
 Output retrieved alignments to a file 
 .I <f>
-instead of to
-.I stdout.
+instead of to stdout.
 
 .TP
-.BI -O
+.BI \-O
 Output retrieved alignment to a file named
-.I <key>.
+.IR key .
 This is a convenience for saving some typing:
 instead of 
-.B esl-afetch -o RRM_1 msafile RRM_1
+.nf
+ \fB% esl\-afetch \-o RRM_1 msafile RRM_1\fR
+.fi
 you can just type
-.B esl-afetch -O msafile RRM_1.
+.nf
+ \fB% esl\-afetch \-O msafile RRM_1\fR
+.fi
 The
-.B -O 
+.B \-O 
 option only works if you're retrieving a
 single alignment; it is incompatible with 
-.B -f.
+.B \-f.
 
 .TP
-.B --index
+.B \-\-index
 Instead of retrieving a
 .I key,
 the special command
-.B esl-afetch --index
+.B esl\-afetch \-\-index
 .I msafile
 produces an SSI index of the names and accessions
 of the alignments in