gn: start of format command

tools/gn/command_format.cc

Index: tools/gn/command_format.cc
diff --git a/tools/gn/command_format.cc b/tools/gn/command_format.cc
new file mode 100644
index 0000000000000000000000000000000000000000..117700284b5977c6932c04ac124b894d32d2f424
--- /dev/null
+++ b/tools/gn/command_format.cc
@@ -0,0 +1,382 @@
+// Copyright 2014 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include <sstream>
+
+#include "base/command_line.h"
+#include "tools/gn/commands.h"
+#include "tools/gn/input_file.h"
+#include "tools/gn/parser.h"
+#include "tools/gn/scheduler.h"
+#include "tools/gn/setup.h"
+#include "tools/gn/source_file.h"
+#include "tools/gn/tokenizer.h"
+
+namespace commands {
+
+const char kSwitchDumpTree[] = "dump-tree";
+
+const char kFormat[] = "format";
+const char kFormat_HelpShort[] =
+    "format: Format .gn file.";
+const char kFormat_Help[] =
+    "gn format: Format .gn file.\n"

[brettw, 2014/09/25 22:11:32]

Can you append here "(alpha, use at own risk)" or something. It's easy to see
the commands and not read the long help.

[scottmg, 2014/09/25 23:03:11]

Done.

+    "\n"
+    "  gn format //some/BUILD.gn\n"
+    "  gn format some\\BUILD.gn\n"
+    "\n"
+    "  Formats .gn file to a standard format. THIS IS NOT FULLY IMPLEMENTED\n"
+    "  YET! IT WILL EAT YOUR BEAUTIFUL .GN FILES. AND YOUR LAUNDRY.\n"
+    "  At a minimum, make sure everything is `git commit`d so you can\n"
+    "  `git checkout -f` to recover.\n";
+
+namespace {
+
+class Printer {
+ public:
+  Printer();
+  ~Printer();
+
+  void Block(const ParseNode* file);
+
+  std::string String() const { return output_; }
+
+ private:
+  // Add to output.
+  void Print(base::StringPiece str);
+
+  // Formatted output.
+  void Printf(const char* fmt, ...);
+
+  void TrimAndPrintToken(const Token& token);
+
+  // End the current line, flushing end of line comments.
+  void Newline();
+
+  // Remove trailing spaces from the current line.
+  void Trim();
+
+  // Get the 0-based x position on the current line.
+  int CurrentColumn();
+
+  enum ExprStyle {
+    kExprStyleRegular,
+    kExprStyleComment,
+  };
+  // Print the expression to the output buffer. Returns the type of element
+  // added to the output.
+  ExprStyle Expr(const ParseNode* root);
+
+  // Format a list of values using the given style.

[brettw, 2014/09/25 22:11:32]

These enums should go at the top of the "private" section.

[scottmg, 2014/09/25 23:03:11]

Done.

+  enum SequenceStyle {
+    kSequenceStyleFunctionCall,
+    kSequenceStyleList,
+    kSequenceStyleBlock,
+  };
+
+  template <class PARSENODE>  // Just for const covariance.
+  void Sequence(SequenceStyle style, const std::vector<PARSENODE*>& list);
+
+  enum { kIndentSize = 2 };

[brettw, 2014/09/25 22:11:32]

Can this be a static const int instead so it's typed?

[scottmg, 2014/09/25 23:03:11]

Done.

+
+  std::string output_;           // Output buffer.
+  std::vector<Token> comments_;  // Pending end-of-line comments.
+  int margin_;                   // Left margin (number of spaces).
+
+  DISALLOW_COPY_AND_ASSIGN(Printer);
+};
+
+Printer::Printer() : margin_(0) {
+  output_.reserve(100 << 10);
+}
+
+Printer::~Printer() {
+}
+
+void Printer::Print(base::StringPiece str) {
+  str.AppendToString(&output_);
+}
+
+void Printer::Printf(const char* fmt, ...) {
+  va_list ap;
+  va_start(ap, fmt);
+  char buf[256];  // Intended for single line output, so shouldn't be a factor.
+  vsnprintf(buf, sizeof(buf), fmt, ap);
+  va_end(ap);
+  output_ += buf;

[brettw, 2014/09/25 22:11:32]

Can this all be replaced with:
  va_list ap;
  va_start(ap, fmt);
  output_ += base::StringPrintV(fmt, ap);
  va_end(ap);

Actually, I take that back. It looks like the only thing this is used for is
printing margins and I find the usage difficult to read. Maybe add a new
function PrintMargin that creates a std::string with the right number of spaces
and prints it, and then use plain Print() for the non-margin chars?

[scottmg, 2014/09/25 23:03:11]

Good point! I had other more complicated format strings before, but they got
simpler.

Done.

+}
+
+void Printer::TrimAndPrintToken(const Token& token) {
+  std::string trimmed;
+  TrimWhitespaceASCII(token.value().as_string(), base::TRIM_ALL, &trimmed);
+  Print(trimmed);
+}
+
+void Printer::Newline() {
+  if (!comments_.empty()) {
+    Print("  ");
+    int i = 0;
+    for (const auto& c : comments_) {
+      if (i > 0) {
+        Trim();
+        Printf("\n%*s", margin_, "");
+      }
+      TrimAndPrintToken(c);
+    }
+    comments_.clear();
+  }
+  Trim();
+  Printf("\n%*s", margin_, "");
+}
+
+void Printer::Trim() {
+  size_t n = output_.size();
+  while (n > 0 && output_[n - 1] == ' ')
+    --n;
+  output_.resize(n);
+}
+
+int Printer::CurrentColumn() {
+  int n = 0;
+  while (n < static_cast<int>(output_.size()) &&
+         output_[output_.size() - 1 - n] != '\n') {
+    ++n;
+  }
+  return n;
+}
+
+void Printer::Block(const ParseNode* root) {
+  const BlockNode* block = root->AsBlock();
+
+  if (block->comments()) {
+    for (const auto& c : block->comments()->before()) {
+      TrimAndPrintToken(c);
+      Newline();
+    }
+  }
+
+  size_t i = 0;
+  for (const auto& stmt : block->statements()) {
+    Expr(stmt);
+    Newline();
+    if (stmt->comments()) {

[brettw, 2014/09/25 22:11:32]

What about "before" statement comments?

[scottmg, 2014/09/25 23:03:11]

Hmm, the asymmetry is kind of confusing. The before (and suffix) are handled
inside Expr() with before() directly printed, and suffix() queued to the next
newline (so that they go after trailing commas). But after() should follow the
newline. But since it's a general expression printer, the newline can't happen
inside, because there's only newlines for statements in blocks. So the after()s
are here.

But what about Sequence then, it should need to print the after() comments in
the same way? I don't think there's any case (at least at the moment) where
after()s are assigned in the middle of a list. They're either standalone, or
attached as before() to a list element. Added CHECK for that.

[brettw, 2014/09/25 23:06:58]

Can you explain this in a comment?

[scottmg, 2014/09/25 23:13:12]

Done.

+      for (const auto& c : stmt->comments()->after()) {
+        TrimAndPrintToken(c);
+        Newline();
+      }
+    }
+    if (i < block->statements().size() - 1)
+      Newline();
+    ++i;
+  }
+
+  if (block->comments()) {
+    for (const auto& c : block->comments()->after()) {
+      TrimAndPrintToken(c);
+      Newline();
+    }
+  }
+}
+
+Printer::ExprStyle Printer::Expr(const ParseNode* root) {
+  ExprStyle result = kExprStyleRegular;
+  if (root->comments()) {
+    if (!root->comments()->before().empty()) {
+      Trim();
+      // If there's already other text on the line, start a new line.
+      if (CurrentColumn() > 0)
+        Print("\n");
+      // We're printing a line comment, so we need to be at the current margin.
+      Printf("%*s", margin_, "");
+      for (const auto& c : root->comments()->before()) {
+        TrimAndPrintToken(c);
+        Newline();
+      }
+    }
+  }
+
+  if (const AccessorNode* accessor = root->AsAccessor()) {
+    Print("TODO(scottmg): AccessorNode");
+  } else if (const BinaryOpNode* binop = root->AsBinaryOp()) {
+    // TODO(scottmg): Lots to do here for complex if expressions: reflowing,
+    // parenthesizing, etc.
+    Expr(binop->left());
+    Print(" ");
+    Print(binop->op().value());
+    Print(" ");
+    Expr(binop->right());
+  } else if (const BlockNode* block = root->AsBlock()) {
+    Sequence(kSequenceStyleBlock, block->statements());
+  } else if (const ConditionNode* condition = root->AsConditionNode()) {
+    Print("if (");
+    Expr(condition->condition());
+    Print(") {");
+    margin_ += kIndentSize;
+    Newline();
+    Block(condition->if_true());
+    margin_ -= kIndentSize;
+    Trim();
+    Printf("%*s}", margin_, "");
+    if (condition->if_false()) {
+      Print(" else ");
+      // If it's a block it's a bare 'else', otherwise it's an 'else if'. See
+      // ConditionNode::Execute.
+      bool is_else_if = condition->if_false()->AsBlock() == NULL;
+      if (is_else_if) {
+        Expr(condition->if_false());
+      } else {
+        Print("{");
+        margin_ += kIndentSize;
+        Newline();
+        Block(condition->if_false());
+        margin_ -= kIndentSize;
+        Trim();
+        Printf("%*s}", margin_, "");
+      }
+    }
+  } else if (const FunctionCallNode* func_call = root->AsFunctionCall()) {
+    Print(func_call->function().value());
+    Sequence(kSequenceStyleFunctionCall, func_call->args()->contents());
+    Print(" {");
+    margin_ += kIndentSize;
+    Newline();
+    Block(func_call->block());
+    margin_ -= kIndentSize;
+    Trim();
+    Printf("%*s}", margin_, "");
+  } else if (const IdentifierNode* identifier = root->AsIdentifier()) {
+    Print(identifier->value().value());
+  } else if (const ListNode* list = root->AsList()) {
+    Sequence(kSequenceStyleList, list->contents());
+  } else if (const LiteralNode* literal = root->AsLiteral()) {
+    // TODO(scottmg): Quoting?
+    Print(literal->value().value());
+  } else if (const UnaryOpNode* unaryop = root->AsUnaryOp()) {
+    Print(unaryop->op().value());
+    Expr(unaryop->operand());
+  } else if (const BlockCommentNode* block_comment = root->AsBlockComment()) {
+    Print(block_comment->comment().value());
+    result = kExprStyleComment;
+  } else {
+    CHECK(false) << "Unhandled case in Expr.";
+  }
+
+  // Defer any end of line comment until we reach the newline.
+  if (root->comments() && !root->comments()->suffix().empty()) {
+    std::copy(root->comments()->suffix().begin(),
+              root->comments()->suffix().end(),
+              std::back_inserter(comments_));
+  }
+
+  return result;
+}
+
+template <class PARSENODE>
+void Printer::Sequence(SequenceStyle style,
+                       const std::vector<PARSENODE*>& list) {
+  bool force_multiline = false;
+  if (style == kSequenceStyleFunctionCall)
+    Print("(");
+  else if (style == kSequenceStyleList)
+    Print("[");
+
+  if (style == kSequenceStyleBlock)
+    force_multiline = true;
+
+  // If there's before line comments, make sure we have a place to put them.
+  for (const auto& i : list) {
+    if (i->comments() && !i->comments()->before().empty())
+      force_multiline = true;
+  }
+
+  if (list.size() == 0 && !force_multiline) {
+    // No elements, and not forcing newlines, print nothing.
+  } else if (list.size() == 1 && !force_multiline) {
+    if (style != kSequenceStyleFunctionCall)
+      Print(" ");
+    Expr(list[0]);
+    if (style != kSequenceStyleFunctionCall)
+      Print(" ");
+  } else {
+    margin_ += kIndentSize;
+    size_t i = 0;
+    for (const auto& x : list) {
+      Newline();
+      ExprStyle expr_style = Expr(x);
+      if (i < list.size() - 1 || style == kSequenceStyleList) {
+        if (expr_style == kExprStyleRegular)
+          Print(",");
+        else
+          Newline();
+      }
+      ++i;
+    }
+
+    margin_ -= kIndentSize;
+    Newline();
+  }
+
+  if (style == kSequenceStyleFunctionCall)
+    Print(")");
+  else if (style == kSequenceStyleList)
+    Print("]");
+}
+
+}  // namespace
+
+bool FormatFileToString(const std::string& input_filename,
+                        std::string* output, bool dump_tree) {

[brettw, 2014/09/25 22:11:32]

Out arg should be last.

[scottmg, 2014/09/25 23:03:11]

Done.

+  Setup setup;
+  Err err;
+  SourceFile input_file(input_filename);
+  const ParseNode* parse_node =
+      setup.scheduler().input_file_manager()->SyncLoadFile(
+          LocationRange(), &setup.build_settings(), input_file, &err);
+  if (err.has_error()) {
+    err.PrintToStdout();
+    return false;
+  }
+  if (dump_tree) {
+    std::ostringstream os;
+    parse_node->Print(os, 0);
+    printf("----------------------\n");
+    printf("-- PARSE TREE --------\n");
+    printf("----------------------\n");
+    printf("%s", os.str().c_str());
+    printf("----------------------\n");
+  }
+  Printer pr;
+  pr.Block(parse_node);
+  *output = pr.String();
+  return true;
+}
+
+int RunFormat(const std::vector<std::string>& args) {
+  // TODO(scottmg): Eventually, this should be a list/spec of files, and they
+  // should all be done in parallel and in-place. For now, we don't want to
+  // overwrite good data with mistakenly reformatted stuff, so we just simply
+  // print the formatted output to stdout.
+  if (args.size() != 1) {
+    Err(Location(), "Expecting exactly one argument, see `gn help format`.\n")
+        .PrintToStdout();
+    return 1;
+  }
+
+  bool dump_tree =
+      base::CommandLine::ForCurrentProcess()->HasSwitch(kSwitchDumpTree);
+
+  std::string input_name = args[0];
+  if (input_name[0] != '/') {
+    std::replace(input_name.begin(), input_name.end(), '\\', '/');
+    input_name = "//" + input_name;
+  }
+  std::string output_string;
+  if (FormatFileToString(input_name, &output_string, dump_tree)) {
+    printf("%s", output_string.c_str());
+  }
+
+  return 0;
+}
+
+}  // namespace commands