Merge pull request #28923 from Kumataro:safeHandlingCompressedFileStorage

core: safe handling of compressed file level extension (.gz[0-9])

modules/core/include/opencv2/core/persistence.hpp

@@ -306,8 +306,12 @@ public:
      before opening the file.
      @param filename Name of the file to open or the text string to read the data from.
      Extension of the file (.xml, .yml/.yaml or .json) determines its format (XML, YAML or JSON
-     respectively). Also you can append .gz to work with compressed files, for example myHugeMatrix.xml.gz. If both
-     FileStorage::WRITE and FileStorage::MEMORY flags are specified, source is used just to specify
+     respectively). Also you can append .gz to work with compressed files, for example myHugeMatrix.xml.gz.
+     You can also specify a compression level from 0 to 9 by appending it to the extension
+     (e.g. ".gz0" for no compression, ".gz9" for high compression).
+     The last digit will be truncated internally to write/read.
+     (e.g. If "a.xml.gz9" is specified, "a.xml.gz" is used for the actual file name.)
+     If both FileStorage::WRITE and FileStorage::MEMORY flags are specified, source is used just to specify
      the output file format (e.g. mydata.xml, .yml etc.). A file name can also contain parameters.
      You can use this format, "*?base64" (e.g. "file.json?base64" (case sensitive)), as an alternative to
      FileStorage::BASE64 flag.

modules/core/src/persistence.cpp

@@ -544,18 +544,30 @@ bool FileStorage::Impl::open(const char *filename_or_buf, int _flags, const char
     flags = _flags;
 
     if (!mem_mode) {
-        char *dot_pos = strrchr((char *) filename.c_str(), '.');
+        size_t dot_idx = filename.find_last_of('.');
         char compression = '\0';
 
-        if (dot_pos && dot_pos[1] == 'g' && dot_pos[2] == 'z' &&
-            (dot_pos[3] == '\0' || (cv_isdigit(dot_pos[3]) && dot_pos[4] == '\0'))) {
-            if (append) {
-                CV_Error(cv::Error::StsNotImplemented, "Appending data to compressed file is not implemented");
+        if (dot_idx != std::string::npos)
+        {
+            std::string ext = filename.substr(dot_idx);
+
+            // Instead of `ext.starts_with(".gz")
+            if (ext.size() >= 3 && (ext[1] == 'g') && (ext[2] == 'z') )
+            {
+                if (ext.size() == 3)
+                {
+                    // ".gz"
+                    isGZ = true;
+                }
+
+                else if(ext.size() == 4 && cv_isdigit(ext[3]))
+                {
+                    // ".gz[0-9]"
+                    isGZ = true;
+                    compression = ext[3];
+                    filename.pop_back(); // Replace `gz[0-9]' to 'gz'.
+                }
             }
-            isGZ = true;
-            compression = dot_pos[3];
-            if (compression)
-                dot_pos[3] = '\0';
         }
 
         if (!isGZ) {
@@ -567,6 +579,9 @@ bool FileStorage::Impl::open(const char *filename_or_buf, int _flags, const char
             }
         } else {
 #if USE_ZLIB
+            if (append) {
+                CV_Error(cv::Error::StsNotImplemented, "Appending data to compressed file is not implemented");
+            }
             char mode[] = {write_mode ? 'w' : 'r', 'b', compression ? compression : '3', '\0'};
             gzfile = gzopen(filename.c_str(), mode);
             if (!gzfile)

modules/core/test/test_io.cpp

@@ -2108,9 +2108,19 @@ T fsWriteRead(const T& expectedValue, const char* ext)
     fs_w.release();
 
     FileStorage fs_r(fname, FileStorage::READ);
-
     T value;
     fs_r["value"] >> value;
+    fs_r.release();
+
+    // If ext is `.gz[0-9]`, fname on storage will end with `.gz`.
+    // FileStorage::Impl::open() truncates the last digit internally.
+    if (isdigit(fname.back()))
+    {
+        fname.pop_back();
+    }
+
+    remove(fname.c_str());
+
     return value;
 }
 
@@ -2143,7 +2153,7 @@ TEST_P(FileStorage_exact_type, long_int)
 }
 
 INSTANTIATE_TEST_CASE_P(Core_InputOutput,
-    FileStorage_exact_type, Values(".yml", ".xml", ".json")
+    FileStorage_exact_type, Values(".yml", ".xml", ".json", ".xml.gz", ".xml.gz0", ".xml.gz9")
 );
 
 }} // namespace