Add a ParseURI function and remove the other URI parsing functions.

ParseURI provides all the functionality used by other filesystems right now, so we're able to remove their custom URI parsing code.

- Use ParseURI for URI parsing in GCS and HDFS.
- Note that this fixes a bug where FileSystem::TranslateName("file://foo/bar")
  gives "foo/bar" as the path. "file:///" is the correct prefix in that case.
Change: 134009580

1 files changed, 23 insertions, 24 deletions

diff --git a/tensorflow/core/platform/file_system.cc b/tensorflow/core/platform/file_system.cc
index e168101a05..928bf644f4 100644
--- a/tensorflow/core/platform/file_system.cc
+++ b/tensorflow/core/platform/file_system.cc
@@ -51,35 +51,34 @@ WritableFile::~WritableFile() {}
 
 FileSystemRegistry::~FileSystemRegistry() {}
 
-string GetSchemeFromURI(const string& name) {
-  auto colon_loc = name.find(":");
+void ParseURI(StringPiece remaining, StringPiece* scheme, StringPiece* host,
+              StringPiece* path) {
+  // 0. Parse scheme
   // Make sure scheme matches [a-zA-Z][0-9a-zA-Z.]*
   // TODO(keveman): Allow "+" and "-" in the scheme.
-  if (colon_loc != string::npos &&
-      strings::Scanner(StringPiece(name.data(), colon_loc))
-          .One(strings::Scanner::LETTER)
-          .Many(strings::Scanner::LETTER_DIGIT_DOT)
-          .GetResult()) {
-    return name.substr(0, colon_loc);
+  if (!strings::Scanner(remaining)
+           .One(strings::Scanner::LETTER)
+           .Many(strings::Scanner::LETTER_DIGIT_DOT)
+           .StopCapture()
+           .OneLiteral("://")
+           .GetResult(&remaining, scheme)) {
+    // If there's no scheme, assume the entire string is a path.
+    scheme->clear();
+    host->clear();
+    *path = remaining;
+    return;
   }
-  return "";
-}
 
-string GetNameFromURI(const string& name) {
-  string scheme = GetSchemeFromURI(name);
-  if (scheme == "") {
-    return name;
-  }
-  // Skip the 'scheme:' portion.
-  StringPiece filename{name.data() + scheme.length() + 1,
-                       name.length() - scheme.length() - 1};
-  // If the URI confirmed to scheme://filename, skip the two '/'s and return
-  // filename. Otherwise return the original 'name', and leave it up to the
-  // implementations to handle the full URI.
-  if (filename.Consume("//")) {
-    return filename.ToString();
+  // 1. Parse host
+  if (!strings::Scanner(remaining).ScanUntil('/').GetResult(&remaining, host)) {
+    // No path, so the rest of the URI is the host.
+    *host = remaining;
+    path->clear();
+    return;
   }
-  return name;
+
+  // 2. The rest is the path
+  *path = remaining;
 }
 
 }  // namespace tensorflow