diff options
author | Boris Kolpackov <boris@codesynthesis.com> | 2024-03-19 13:24:44 +0200 |
---|---|---|
committer | Boris Kolpackov <boris@codesynthesis.com> | 2024-03-21 06:44:06 +0200 |
commit | 3a55e033e4fc9a18ede99c4f9dd69fd30c383cf7 (patch) | |
tree | 1c35062021a261479ff5f38d95dd22cfe34fbb38 /libbutl/utility.hxx | |
parent | 736b0f25003c92b3903798ce0a768230480d8f4b (diff) |
In particular, this version can be used to parse lines while observing
blanks.
Diffstat (limited to 'libbutl/utility.hxx')
-rw-r--r-- | libbutl/utility.hxx | 23 |
1 files changed, 23 insertions, 0 deletions
diff --git a/libbutl/utility.hxx b/libbutl/utility.hxx index a129276..779a0aa 100644 --- a/libbutl/utility.hxx +++ b/libbutl/utility.hxx @@ -190,6 +190,24 @@ namespace butl // // The second version examines up to the n'th character in the string. // + // The third version, instead of skipping consecutive delimiters, treats + // them as separating empty words. The additional m variable contains an + // unspecified internal state and should be initialized to 0. Note that in + // this case you should use the (b == n) condition to detect the end. Note + // also that a leading delimiter is considered as separating an empty word + // from the rest and the trailing delimiter is considered as separating the + // rest from an empty word. For example, this is how to parse lines while + // observing blanks: + // + // for (size_t b (0), e (0), m (0), n (s.size ()); + // next_word (s, n, b, e, m, '\n', '\r'), b != n; ) + // { + // string l (s, b, e - b); + // } + // + // For string "\na\n" this code will observe the {"", "a", ""} words. And + // for just "\n" it will observe the {"", ""} words. + // std::size_t next_word (const std::string&, std::size_t& b, std::size_t& e, char d1 = ' ', char d2 = '\0'); @@ -198,6 +216,11 @@ namespace butl next_word (const std::string&, std::size_t n, std::size_t& b, std::size_t& e, char d1 = ' ', char d2 = '\0'); + std::size_t + next_word (const std::string&, std::size_t n, + std::size_t& b, std::size_t& e, std::size_t& m, + char d1 = ' ', char d2 = '\0'); + // Sanitize a string to only contain characters valid in an identifier // (ASCII alphanumeric plus `_`) replacing all others with `_`. // |