Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 1 | //===-- StringExtras.cpp - Implement the StringExtras header --------------===// |
Misha Brukman | f976c85 | 2005-04-21 22:55:34 +0000 | [diff] [blame] | 2 | // |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 3 | // The LLVM Compiler Infrastructure |
| 4 | // |
Chris Lattner | 4ee451d | 2007-12-29 20:36:04 +0000 | [diff] [blame] | 5 | // This file is distributed under the University of Illinois Open Source |
| 6 | // License. See LICENSE.TXT for details. |
Misha Brukman | f976c85 | 2005-04-21 22:55:34 +0000 | [diff] [blame] | 7 | // |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 8 | //===----------------------------------------------------------------------===// |
| 9 | // |
| 10 | // This file implements the StringExtras.h header |
| 11 | // |
| 12 | //===----------------------------------------------------------------------===// |
| 13 | |
Rafael Espindola | c78c0c9 | 2009-11-13 02:18:25 +0000 | [diff] [blame] | 14 | #include "llvm/ADT/SmallVector.h" |
Benjamin Kramer | d4f1959 | 2010-01-11 18:03:24 +0000 | [diff] [blame] | 15 | #include "llvm/ADT/STLExtras.h" |
| 16 | #include "llvm/ADT/StringExtras.h" |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 17 | using namespace llvm; |
| 18 | |
Benjamin Kramer | 7462b5d | 2010-01-11 19:45:18 +0000 | [diff] [blame] | 19 | /// StrInStrNoCase - Portable version of strcasestr. Locates the first |
| 20 | /// occurrence of string 's1' in string 's2', ignoring case. Returns |
| 21 | /// the offset of s2 in s1 or npos if s2 cannot be found. |
| 22 | StringRef::size_type llvm::StrInStrNoCase(StringRef s1, StringRef s2) { |
| 23 | size_t N = s2.size(), M = s1.size(); |
| 24 | if (N > M) |
| 25 | return StringRef::npos; |
| 26 | for (size_t i = 0, e = M - N + 1; i != e; ++i) |
| 27 | if (s1.substr(i, N).equals_lower(s2)) |
| 28 | return i; |
| 29 | return StringRef::npos; |
| 30 | } |
| 31 | |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 32 | /// getToken - This function extracts one token from source, ignoring any |
| 33 | /// leading characters that appear in the Delimiters string, and ending the |
| 34 | /// token at any of the characters that appear in the Delimiters string. If |
| 35 | /// there are no tokens in the source string, an empty string is returned. |
Benjamin Kramer | d4f1959 | 2010-01-11 18:03:24 +0000 | [diff] [blame] | 36 | /// The function returns a pair containing the extracted token and the |
| 37 | /// remaining tail string. |
| 38 | std::pair<StringRef, StringRef> llvm::getToken(StringRef Source, |
| 39 | StringRef Delimiters) { |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 40 | // Figure out where the token starts. |
Benjamin Kramer | d4f1959 | 2010-01-11 18:03:24 +0000 | [diff] [blame] | 41 | StringRef::size_type Start = Source.find_first_not_of(Delimiters); |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 42 | |
Benjamin Kramer | d4f1959 | 2010-01-11 18:03:24 +0000 | [diff] [blame] | 43 | // Find the next occurrence of the delimiter. |
| 44 | StringRef::size_type End = Source.find_first_of(Delimiters, Start); |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 45 | |
Benjamin Kramer | 4e7e2f5 | 2010-01-18 12:40:05 +0000 | [diff] [blame] | 46 | return std::make_pair(Source.slice(Start, End), Source.substr(End)); |
Chris Lattner | 872ccce | 2003-12-29 05:07:02 +0000 | [diff] [blame] | 47 | } |
Chris Lattner | 53eca94 | 2006-07-14 22:54:39 +0000 | [diff] [blame] | 48 | |
Chris Lattner | d5b58c2 | 2006-11-28 22:32:35 +0000 | [diff] [blame] | 49 | /// SplitString - Split up the specified string according to the specified |
| 50 | /// delimiters, appending the result fragments to the output list. |
Benjamin Kramer | d4f1959 | 2010-01-11 18:03:24 +0000 | [diff] [blame] | 51 | void llvm::SplitString(StringRef Source, |
| 52 | SmallVectorImpl<StringRef> &OutFragments, |
| 53 | StringRef Delimiters) { |
| 54 | StringRef S2, S; |
| 55 | tie(S2, S) = getToken(Source, Delimiters); |
Chris Lattner | d5b58c2 | 2006-11-28 22:32:35 +0000 | [diff] [blame] | 56 | while (!S2.empty()) { |
| 57 | OutFragments.push_back(S2); |
Benjamin Kramer | d4f1959 | 2010-01-11 18:03:24 +0000 | [diff] [blame] | 58 | tie(S2, S) = getToken(S, Delimiters); |
Chris Lattner | d5b58c2 | 2006-11-28 22:32:35 +0000 | [diff] [blame] | 59 | } |
| 60 | } |
Rafael Espindola | 5ccac24 | 2009-11-13 01:24:40 +0000 | [diff] [blame] | 61 | |
Rafael Espindola | c78c0c9 | 2009-11-13 02:18:25 +0000 | [diff] [blame] | 62 | void llvm::StringRef::split(SmallVectorImpl<StringRef> &A, |
| 63 | StringRef Separators, int MaxSplit, |
Rafael Espindola | 5ccac24 | 2009-11-13 01:24:40 +0000 | [diff] [blame] | 64 | bool KeepEmpty) const { |
| 65 | StringRef rest = *this; |
| 66 | |
Rafael Espindola | 20fd4ec | 2009-11-13 04:55:09 +0000 | [diff] [blame] | 67 | // rest.data() is used to distinguish cases like "a," that splits into |
| 68 | // "a" + "" and "a" that splits into "a" + 0. |
Rafael Espindola | c78c0c9 | 2009-11-13 02:18:25 +0000 | [diff] [blame] | 69 | for (int splits = 0; |
Rafael Espindola | 20fd4ec | 2009-11-13 04:55:09 +0000 | [diff] [blame] | 70 | rest.data() != NULL && (MaxSplit < 0 || splits < MaxSplit); |
Rafael Espindola | 5ccac24 | 2009-11-13 01:24:40 +0000 | [diff] [blame] | 71 | ++splits) { |
| 72 | std::pair<llvm::StringRef, llvm::StringRef> p = rest.split(Separators); |
| 73 | |
| 74 | if (p.first.size() != 0 || KeepEmpty) |
| 75 | A.push_back(p.first); |
| 76 | rest = p.second; |
| 77 | } |
Rafael Espindola | 20fd4ec | 2009-11-13 04:55:09 +0000 | [diff] [blame] | 78 | // If we have a tail left, add it. |
| 79 | if (rest.data() != NULL && (rest.size() != 0 || KeepEmpty)) |
Rafael Espindola | 5ccac24 | 2009-11-13 01:24:40 +0000 | [diff] [blame] | 80 | A.push_back(rest); |
| 81 | } |