Simplify INI-style config reader using C++11 STL (#4478)

* simplify the config.h file

* revise config.h

* revised config.h

* revise format

* revise format issues

* revise whitespace issues

* revise whitespace namespace format issues

* revise namespace format issues

* format issues

* format issues

* format issues

* format issues

* Revert submodule changes

* minor change

* Update src/common/config.h

Co-Authored-By: Philip Hyunsu Cho <chohyu01@cs.washington.edu>

* address format issue from trivialfis

* Use correct cub submodule

CONTRIBUTORS.md

@@ -91,3 +91,4 @@ List of Contributors
 * [Jiaxiang Li](https://github.com/JiaxiangBU)
 * [Bryan Woods](https://github.com/bryan-woods)
   - Bryan added support for cross-validation for the ranking objective
+* [Haoda Fu](https://github.com/fuhaoda)

src/cli_main.cc

@@ -341,13 +341,10 @@ int CLIRunTask(int argc, char *argv[]) {
   }
   rabit::Init(argc, argv);
 
-  std::vector<std::pair<std::string, std::string> > cfg;
+  common::ConfigParse cp(argv[1]);
+  auto cfg = cp.Parse();
   cfg.emplace_back("seed", "0");
 
-  common::ConfigIterator itr(argv[1]);
-  while (itr.Next()) {
-    cfg.emplace_back(std::string(itr.Name()), std::string(itr.Val()));
-  }
 
   for (int i = 2; i < argc; ++i) {
     char name[256], val[256];

src/common/config.h

@@ -1,8 +1,8 @@
 /*!
- * Copyright 2014 by Contributors
+ * Copyright 2014-2019 by Contributors
  * \file config.h
  * \brief helper class to load in configures from file
- * \author Tianqi Chen
+ * \author Haoda Fu
  */
 #ifndef XGBOOST_COMMON_CONFIG_H_
 #define XGBOOST_COMMON_CONFIG_H_
@@ -12,181 +12,73 @@
 #include <string>
 #include <istream>
 #include <fstream>
+#include <vector>
+#include <utility>
 
 namespace xgboost {
 namespace common {
 /*!
- * \brief base implementation of config reader
+ * \brief Implementation of config reader
  */
-class ConfigReaderBase {
+class ConfigParse {
- public:
-  /*!
-   * \brief get current name, called after Next returns true
-   * \return current parameter name
-   */
-  inline const char *Name() const {
-    return s_name_.c_str();
-  }
-  /*!
-   * \brief get current value, called after Next returns true
-   * \return current parameter value
-   */
-  inline const char *Val() const {
-    return s_val_.c_str();
-  }
-  /*!
-   * \brief move iterator to next position
-   * \return true if there is value in next position
-   */
-  inline bool Next() {
-    while (!this->IsEnd()) {
-      GetNextToken(&s_name_);
-      if (s_name_ == "=") return false;
-      if (GetNextToken(&s_buf_) || s_buf_ != "=")  return false;
-      if (GetNextToken(&s_val_) || s_val_ == "=")  return false;
-      return true;
-    }
-    return false;
-  }
-  // called before usage
-  inline void Init() {
-    ch_buf_ = this->GetChar();
-  }
-
- protected:
-  /*!
-   * \brief to be implemented by subclass,
-   * get next token, return EOF if end of file
-   */
-  virtual int GetChar() = 0;
-  /*! \brief to be implemented by child, check if end of stream */
-  virtual bool IsEnd() = 0;
-
- private:
-  int ch_buf_;
-  std::string s_name_, s_val_, s_buf_;
-
-  inline void SkipLine() {
-    do {
-      ch_buf_ = this->GetChar();
-    } while (ch_buf_ != EOF && ch_buf_ != '\n' && ch_buf_ != '\r');
-  }
-
-  inline void ParseStr(std::string *tok) {
-    while ((ch_buf_ = this->GetChar()) != EOF) {
-      switch (ch_buf_) {
-        case '\\': *tok += this->GetChar(); break;
-        case '\"': return;
-        case '\r':
-        case '\n': LOG(FATAL)<< "ConfigReader: unterminated string";
-        default: *tok += static_cast<char>(ch_buf_);
-      }
-    }
-    LOG(FATAL) << "ConfigReader: unterminated string";
-  }
-  inline void ParseStrML(std::string *tok) {
-    while ((ch_buf_ = this->GetChar()) != EOF) {
-      switch (ch_buf_) {
-        case '\\': *tok += this->GetChar(); break;
-        case '\'': return;
-        default: *tok += static_cast<char>(ch_buf_);
-      }
-    }
-    LOG(FATAL) << "unterminated string";
-  }
-  // return newline
-  inline bool GetNextToken(std::string *tok) {
-    tok->clear();
-    bool new_line = false;
-    while (ch_buf_ != EOF) {
-      switch (ch_buf_) {
-        case '#' : SkipLine(); new_line = true; break;
-        case '\"':
-          if (tok->length() == 0) {
-            ParseStr(tok); ch_buf_ = this->GetChar(); return new_line;
-          } else {
-            LOG(FATAL) << "ConfigReader: token followed directly by string";
-          }
-        case '\'':
-          if (tok->length() == 0) {
-            ParseStrML(tok); ch_buf_ = this->GetChar(); return new_line;
-          } else {
-            LOG(FATAL) << "ConfigReader: token followed directly by string";
-          }
-        case '=':
-          if (tok->length() == 0) {
-            ch_buf_ = this->GetChar();
-            *tok = '=';
-          }
-          return new_line;
-        case '\r':
-        case '\n':
-          if (tok->length() == 0) new_line = true;
-        case '\t':
-        case ' ' :
-          ch_buf_ = this->GetChar();
-          if (tok->length() != 0) return new_line;
-          break;
-        default:
-          *tok += static_cast<char>(ch_buf_);
-          ch_buf_ = this->GetChar();
-          break;
-      }
-    }
-    if (tok->length() == 0) {
-      return true;
-    } else {
-      return false;
-    }
-  }
-};
-/*!
- * \brief an iterator use stream base, allows use all types of istream
- */
-class ConfigStreamReader: public ConfigReaderBase {
  public:
   /*!
    * \brief constructor
-   * \param fin istream input stream
+   * \param cfgFileName name of configure file
    */
-  explicit ConfigStreamReader(std::istream &fin) : fin_(fin) {}
+  explicit ConfigParse(const std::string &cfgFileName) {
-
+    fi_.open(cfgFileName);
- protected:
-  int GetChar() override {
-    return fin_.get();
-  }
-  /*! \brief to be implemented by child, check if end of stream */
-  bool IsEnd() override {
-    return fin_.eof();
-  }
-
- private:
-  std::istream &fin_;
-};
-
-/*!
- * \brief an iterator that iterates over a configure file and gets the configures
- */
-class ConfigIterator: public ConfigStreamReader {
- public:
-  /*!
-   * \brief constructor
-   * \param fname name of configure file
-   */
-  explicit ConfigIterator(const char *fname) : ConfigStreamReader(fi_) {
-    fi_.open(fname);
     if (fi_.fail()) {
-      LOG(FATAL) << "cannot open file " << fname;
+      LOG(FATAL) << "cannot open file " << cfgFileName;
     }
-    ConfigReaderBase::Init();
   }
-  /*! \brief destructor */
+
-  ~ConfigIterator() {
+  /*!
+   * \brief parse the configure file
+   */
+  std::vector<std::pair<std::string, std::string> > Parse() {
+    std::vector<std::pair<std::string, std::string> > results{};
+    char delimiter = '=';
+    char comment = '#';
+    std::string line{};
+    std::string name{};
+    std::string value{};
+
+    while (!fi_.eof()) {
+      std::getline(fi_, line);  // read a line of configure file
+      line = line.substr(0, line.find(comment));  // anything beyond # is comment
+      size_t delimiterPos = line.find(delimiter);  // find the = sign
+      name = line.substr(0, delimiterPos);  // anything before = is the name
+      // after this = is the value
+      value = line.substr(delimiterPos + 1, line.length() - delimiterPos - 1);
+
+      if (line.empty() || name.empty() || value.empty())
+        continue;  // skip a line if # at beginning or there is no value or no name.
+      CleanString(&name);  // clean the string
+      CleanString(&value);
+      results.emplace_back(name, value);
+    }
+    return results;
+  }
+
+  ~ConfigParse() {
     fi_.close();
   }
 
  private:
   std::ifstream fi_;
+  std::string allowableChar_ =
+      "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_-./\\";
+
+  /*!
+   * \brief remove unnecessary chars.
+   */
+  void CleanString(std::string * str) {
+    size_t firstIndx = str->find_first_of(allowableChar_);
+    size_t lastIndx = str->find_last_of(allowableChar_);
+    // this line can be more efficient, but keep as is for simplicity.
+    *str = str->substr(firstIndx, lastIndx - firstIndx + 1);
+  }
 };
 }  // namespace common
 }  // namespace xgboost