updated to change uint32 to uint64 for file size and disabled EOF req…

…uirement for gz files (#87)
HumanCellAtlas · Nov 3, 2020 · 65ac1a0 · 65ac1a0
1 parent 4813538
commit 65ac1a0
Show file tree

Hide file tree

Showing 4 changed files with 19 additions and 7 deletions.
diff --git a/Dockerfile b/Dockerfile
@@ -24,6 +24,7 @@ RUN cd /sctools/fastqpreprocessing &&\
     tar -zxvf v${libStatGen_version}.tar.gz &&\
     mv libStatGen-${libStatGen_version} libStatGen &&\
     patch libStatGen/fastq/FastQFile.cpp patches/FastQFile.cpp.patch &&\
+    patch libStatGen/general/BgzfFileType.cpp patches/BgzfFileType.cpp.patch &&\  
     patch libStatGen/Makefile patches/Makefile.patch &&\
     patch libStatGen/general/Makefile patches/general.Makefile.patch &&\
     make -C libStatGen &&\

diff --git a/fastqpreprocessing/patches/BgzfFileType.cpp.patch b/fastqpreprocessing/patches/BgzfFileType.cpp.patch
@@ -0,0 +1,11 @@
+--- libStatGen/general/BgzfFileType.cpp	2015-07-08 20:03:23.000000000 +0000
++++ /tmp/BgzfFileType.cpp	2020-11-03 12:25:36.168474179 +0000
+@@ -23,7 +23,7 @@
+ #include "BgzfFileType.h"
+
+ // Default to require the EOF block at the end of the file.
+-bool BgzfFileType::ourRequireEofBlock = true;
++bool BgzfFileType::ourRequireEofBlock = false;
+
+ BgzfFileType::BgzfFileType(const char * filename, const char * mode)
+ {
diff --git a/fastqpreprocessing/src/utilities.cpp b/fastqpreprocessing/src/utilities.cpp
@@ -14,10 +14,10 @@ using namespace std;
 namespace fs = std::experimental::filesystem;
 
 /** @copydoc filesize */
-int32_t filesize(const char *filename) {
+int64_t filesize(const char *filename) {
     FILE *f = fopen(filename, "rb");  /* open the file in read only */
 
-    int32_t size = 0;
+    int64_t size = 0;
     if (fseek(f, 0, SEEK_END) ==0 ) /* seek was successful */
        size = ftell(f);
     fclose(f);
@@ -286,7 +286,7 @@ void _print_file_info(const std::vector<std::string> &fastqs,
 
 
 /** @copydoc  get_num_blocks */
-int32_t get_num_blocks(const INPUT_OPTIONS &options) {
+int64_t get_num_blocks(const INPUT_OPTIONS &options) {
     double tot_size = 0;
     for (int i= 0; i < options.R1s.size(); i++) {
         if (options.I1s.size()) {

diff --git a/fastqpreprocessing/src/utilities.h b/fastqpreprocessing/src/utilities.h
@@ -19,7 +19,7 @@ typedef std::pair<std::string, bool>  STRING_BOOL_PAIR;
 
 typedef std::vector<std::string>  STRING_VECTOR;
 
-typedef std::unordered_map <std::string, int32_t> STRING_INT32_MAP;
+typedef std::unordered_map <std::string, int64_t> STRING_INT32_MAP;
 
 // structure for correcting the barcodes
 typedef struct _white_list_data {
@@ -61,7 +61,7 @@ typedef struct _input_options {
 
  * @param options Input options structure that contains file name
 */
-int32_t get_num_blocks(const INPUT_OPTIONS &options);
+int64_t get_num_blocks(const INPUT_OPTIONS &options);
 
 /**
  * @brief Build barcode correction map white list barcodes & mutations
@@ -95,7 +95,7 @@ void read_options(int, char **, INPUT_OPTIONS &);
  *  @param filename file name whose size is computed
  *  @return size of the file in bytes
 */
-int32_t filesize(const char *filename);
+int64_t filesize(const char *filename);
 
 
 /**
@@ -104,7 +104,7 @@ int32_t filesize(const char *filename);
  *  @param filename file name whose size is computed
  *  @return size of the file in bytes
 */
-int32_t getFileSize(const std::string &fileName);
+int64_t getFileSize(const std::string &fileName);
 
 /**
  * @brief Print system error and exit