Add Windows compatibility, geo_loc_name support, and fix memory leaks

- Add support for /geo_loc_name qualifier (GenBank 2024 change) in addition to /country
- Add Windows compatibility with portable strptime and getopt implementations
- Replace VLA with dynamic allocation for MSVC compatibility
- Fix memory leak: sCountry2 was allocated but never freed in main loop
- Fix memory leak: add freeSingleGBData() for proper cleanup on parse errors
- Update README with Windows build instructions (WSL, MSYS2, PCRE2, TRE)

Author: Simon Frost

README.md

@@ -9,7 +9,7 @@ This little C program will extract the following information from a GenBank file
 - length
 - submission date
 - host
-- country
+- country (supports both `/country` and `/geo_loc_name` qualifiers)
 - collection date
 
 In addition to extracting this information, dates are reformatted e.g. `31-DEC-2001` becomes `2001-12-31`, which makes them more digestible to downstream software like BEAST, and country names are cleaned and matched to ISO3 codes.
@@ -30,6 +30,8 @@ gbmunge [-h] -i <Genbank_file> -f <sequence_output> -o <metadata_output> [-t] [-
 
 ## Building
 
+### Linux and macOS
+
 ```sh
 git clone https://github.com/sdwfrost/gbmunge
 cd gbmunge
@@ -38,6 +40,40 @@ make
 
 This will build `gbmunge` in the `src/` directory. Add the directory to the path, or move the executable somewhere.
 
+### Windows
+
+There are several options for building on Windows:
+
+1. **Using WSL (Windows Subsystem for Linux)** (Recommended):
+   ```sh
+   # Install WSL with Ubuntu, then in the WSL terminal:
+   sudo apt update
+   sudo apt install build-essential
+   cd gbmunge
+   make
+   ```
+
+2. **Using MSYS2/MinGW**:
+   ```sh
+   # Install MSYS2, then in MSYS2 terminal:
+   pacman -S mingw-w64-x86_64-gcc make
+   cd gbmunge
+   make
+   ```
+
+3. **Using Visual Studio with vcpkg**:
+   Building natively with MSVC requires a POSIX-compatible regex library:
+   ```sh
+   # Install PCRE2 via vcpkg
+   vcpkg install pcre2:x64-windows
+   
+   # Compile with PCRE2 support (modify Makefile or compile manually)
+   cl /DGBMUNGE_USE_PCRE2 /I<vcpkg_include_path> gbfp.c gbmunge.c /link pcre2-8.lib
+   ```
+
+4. **Using TRE regex library**:
+   Download TRE from https://github.com/laurikari/tre and place the POSIX-compatible `regex.h` in the `src/` directory.
+
 ## Testing
 
 A Genbank file of MERS Coronavirus sequences is provided in the `test/` directory.

src/gbfp.c

@@ -2,12 +2,27 @@
 
 #include <stdio.h>
 #include <stdlib.h>
-#include <regex.h>
-#include <unistd.h>
 #include <string.h>
 #include <ctype.h>
 #include <limits.h>
-#include <sys/types.h>
+
+#ifdef _WIN32
+    #include <windows.h>
+    /* For Windows, use PCRE2 or bundled regex - user must install PCRE2 */
+    /* Alternatively, define GBMUNGE_USE_PCRE2 and link with pcre2 */
+    #ifdef GBMUNGE_USE_PCRE2
+        #define PCRE2_CODE_UNIT_WIDTH 8
+        #include <pcre2.h>
+        /* PCRE2 wrapper - would need implementation */
+    #else
+        /* Fallback: use bundled regex implementation or TRE */
+        #include "regex.h"  /* User must provide POSIX-compatible regex.h for Windows */
+    #endif
+#else
+    #include <regex.h>
+    #include <unistd.h>
+    #include <sys/types.h>
+#endif
 
 #include "gbfp.h"
 
@@ -55,6 +70,9 @@ void freeRegEx(void) {
     regfree(&ptRegExGI);
 }
 
+/* Forward declaration for freeSingleGBData */
+static void freeSingleGBData(gb_data *ptGBData);
+
 /* Removes white spaces at end of a string */
 static void rtrim(gb_string sLine) {
     register int i;
@@ -640,7 +658,7 @@ static gb_data *_parseGBFF(FILE *FSeqFile) {
 
     /* Parse LOCUS line */ 
     if (parseLocus(sLine, ptGBData) != 0) {
-        free(ptGBData);
+        freeSingleGBData(ptGBData);
         return NULL;
     }
 
@@ -695,36 +713,35 @@ gb_data **parseGBFF(gb_string spFileName) {
     return pptGBDatas;
 }
 
-void freeGBData(gb_data **pptGBData) {
+/* Free a single gb_data structure and all its contents */
+static void freeSingleGBData(gb_data *ptGBData) {
     unsigned int i;
-    gb_data *ptGBData = NULL;
     gb_feature *ptFeatures = NULL;
     gb_reference *ptReferences = NULL;
     unsigned int iFeatureNum = 0;
     unsigned int iReferenceNum = 0;
-    unsigned int iSeqPos = 0;
-  
-    for (iSeqPos = 0; *(pptGBData + iSeqPos) != NULL; iSeqPos++) {
-        ptGBData = *(pptGBData + iSeqPos);
 
-        ptFeatures = ptGBData->ptFeatures;
-        iFeatureNum = ptGBData->iFeatureNum;
+    if (ptGBData == NULL) return;
 
+    ptFeatures = ptGBData->ptFeatures;
+    iFeatureNum = ptGBData->iFeatureNum;
 
-        /* Release memory space for features */    
+    /* Release memory space for features */    
+    if (ptFeatures != NULL) {
         for (i = 0; i < iFeatureNum; i++) {
             free((ptFeatures + i)->ptLocation);
             if ((ptFeatures + i)->ptQualifier != NULL) {
                 free(((ptFeatures + i)->ptQualifier)->sQualifier);
                 free((ptFeatures + i)->ptQualifier);
             }
         }
-
         free(ptFeatures);
+    }
 
-        /* Release memory space for References */
-        ptReferences = ptGBData->ptReferences;
-        iReferenceNum = ptGBData->iReferenceNum;
+    /* Release memory space for References */
+    ptReferences = ptGBData->ptReferences;
+    iReferenceNum = ptGBData->iReferenceNum;
+    if (ptReferences != NULL) {
         for (i = 0; i < iReferenceNum; i++) {
             free((ptReferences + i)->sAuthors);
             free((ptReferences + i)->sConsrtm);
@@ -735,19 +752,29 @@ void freeGBData(gb_data **pptGBData) {
             free((ptReferences + i)->sRemark);
         }
         free(ptReferences);
+    }
 
-        free(ptGBData->sDef);
-        free(ptGBData->sAccession);
-        free(ptGBData->sComment);
-        free(ptGBData->sGI);
-        free(ptGBData->sKeywords);
-        free(ptGBData->sLineage);
-        free(ptGBData->sOrganism);
-        free(ptGBData->sSequence);
-        free(ptGBData->sSource);
-        free(ptGBData->sVersion);
-
-        free(ptGBData);
+    free(ptGBData->sDef);
+    free(ptGBData->sAccession);
+    free(ptGBData->sComment);
+    free(ptGBData->sGI);
+    free(ptGBData->sKeywords);
+    free(ptGBData->sLineage);
+    free(ptGBData->sOrganism);
+    free(ptGBData->sSequence);
+    free(ptGBData->sSource);
+    free(ptGBData->sVersion);
+
+    free(ptGBData);
+}
+
+void freeGBData(gb_data **pptGBData) {
+    unsigned int iSeqPos = 0;
+  
+    if (pptGBData == NULL) return;
+
+    for (iSeqPos = 0; *(pptGBData + iSeqPos) != NULL; iSeqPos++) {
+        freeSingleGBData(*(pptGBData + iSeqPos));
     }
 
     free(pptGBData);