Problem Creating Regex Expression

Collapse
This topic is closed.
X
X
Collapse
 
  • Page of 1
  • Time
  • Show
Clear All
new posts
Previous template Next
  • Sean
    #1

    Problem Creating Regex Expression

    I am finally taking the time to get to know regex, but it seems I have
    taken a bit of a tumble.

    I have the following (dummy) data:

    <td>Name:</td <td>Kherie Kali</td>

    If I use this expression: "<td>Name:</td>\s*<td>Kheri e Kali</td>"

    I indeed get a match.

    The next step I took is to get this without knowing the name in
    advance. I then used the following expression: <td>Name:</td>
    \s*<td>([a-zA-Z_$][a-zA-Z0-9_$]*)</td>

    and I didn't get a thing.

    Shouldn't the "([a-zA-Z_$][a-zA-Z0-9_$]*)" part match any word or
    symbol and then the * make it multiple words or symbols? Any light
    that could be shed on this situation is appreciated. I don't
    necessarily want the answer to my quandry, but insight as to what I am
    doing wrong.

    Thank you,

    -Sean

    Tags: None
    • Sergey Zyuzin
      #2
      Re: Problem Creating Regex Expression

      On Feb 8, 7:59 am, Sean <ColdFusion...@ gmail.comwrote:
      I am finally taking the time to get to know regex, but it seems I have
      taken a bit of a tumble.
      >
      I have the following (dummy) data:
      >
      <td>Name:</td <td>Kherie Kali</td>
      >
      If I use this expression: "<td>Name:</td>\s*<td>Kheri e Kali</td>"
      >
      I indeed get a match.
      >
      The next step I took is to get this without knowing the name in
      advance. I then used the following expression: <td>Name:</td>
      \s*<td>([a-zA-Z_$][a-zA-Z0-9_$]*)</td>
      >
      and I didn't get a thing.
      >
      Shouldn't the "([a-zA-Z_$][a-zA-Z0-9_$]*)" part match any word or
      symbol and then the * make it multiple words or symbols? Any light
      that could be shed on this situation is appreciated. I don't
      necessarily want the answer to my quandry, but insight as to what I am
      doing wrong.
      >
      Thank you,
      >
      -Sean
      Hi, Sean

      "([a-zA-Z_$][a-zA-Z0-9_$]*)" will match any letter, underscore or '$'
      character followed by zero or more letters, digits, underscores, '$'
      chars.

      It seems you don't take into account space in the middle of "Kherie
      Kali".
      If you write more specific requirements I could write a RegEx

      Thanks,
      Sergey

        Comment

        • Kevin Spencer
          #3
          Re: Problem Creating Regex Expression

          The following will work capture all content inside <td></tdtags:

          (?<=td>)(.*?)(? =<)

          The first part is a positive look-ahead, indicating that a match must be
          preceded by the character sequence "td>" (non-capturing). The second part
          indicates any character 0 or more times with a lazy quantifier, meaning that
          it will capture as few times as possible. The third part is a positive
          look-ahead, indicating that the match must be followed by a "<" character.
          Since there are no "<" characters in the actual tag's content, this stops
          the match at the end of the tag.

          --
          HTH,

          Kevin Spencer
          Chicken Salad Surgeon
          Microsoft MVP

          "Sean" <ColdFusion244@ gmail.comwrote in message
          news:9e791c78-b1a2-43cd-916f-ae0202d31b66@1g 2000hsl.googleg roups.com...
          >I am finally taking the time to get to know regex, but it seems I have
          taken a bit of a tumble.
          >
          I have the following (dummy) data:
          >
          <td>Name:</td <td>Kherie Kali</td>
          >
          If I use this expression: "<td>Name:</td>\s*<td>Kheri e Kali</td>"
          >
          I indeed get a match.
          >
          The next step I took is to get this without knowing the name in
          advance. I then used the following expression: <td>Name:</td>
          \s*<td>([a-zA-Z_$][a-zA-Z0-9_$]*)</td>
          >
          and I didn't get a thing.
          >
          Shouldn't the "([a-zA-Z_$][a-zA-Z0-9_$]*)" part match any word or
          symbol and then the * make it multiple words or symbols? Any light
          that could be shed on this situation is appreciated. I don't
          necessarily want the answer to my quandry, but insight as to what I am
          doing wrong.
          >
          Thank you,
          >
          -Sean
          >

            Comment

            • Sean
              #4
              Re: Problem Creating Regex Expression

              On Feb 8, 8:19 am, "Kevin Spencer" <unclechutney@l ocalhostwrote:
              The following will work capture all content inside <td></tdtags:
              >
              (?<=td>)(.*?)(? =<)
              >
              The first part is a positive look-ahead, indicating that a match must be
              preceded by the character sequence "td>" (non-capturing). The second part
              indicates any character 0 or more times with a lazy quantifier, meaning that
              it will capture as few times as possible. The third part is a positive
              look-ahead, indicating that the match must be followed by a "<" character.
              Since there are no "<" characters in the actual tag's content, this stops
              the match at the end of the tag.
              >
              --
              HTH,
              >
              Kevin Spencer
              Chicken Salad Surgeon
              Microsoft MVP
              >
              "Sean" <ColdFusion...@ gmail.comwrote in message
              >
              news:9e791c78-b1a2-43cd-916f-ae0202d31b66@1g 2000hsl.googleg roups.com...
              >
              >
              >
              I am finally taking the time to get to know regex, but it seems I have
              taken a bit of a tumble.
              >
              I have the following (dummy) data:
              >
              <td>Name:</td <td>Kherie Kali</td>
              >
              If I use this expression: "<td>Name:</td>\s*<td>Kheri e Kali</td>"
              >
              I indeed get a match.
              >
              The next step I took is to get this without knowing the name in
              advance. I then used the following expression: <td>Name:</td>
              \s*<td>([a-zA-Z_$][a-zA-Z0-9_$]*)</td>
              >
              and I didn't get a thing.
              >
              Shouldn't the "([a-zA-Z_$][a-zA-Z0-9_$]*)" part match any word or
              symbol and then the * make it multiple words or symbols? Any light
              that could be shed on this situation is appreciated. I don't
              necessarily want the answer to my quandry, but insight as to what I am
              doing wrong.
              >
              Thank you,
              >
              -Sean- Hide quoted text -
              >
              - Show quoted text -
              Kevin and Sergey,

              Kevin: Thank you for the explanation! This find the Name: Tag, but
              won't find the dummy person's actual name. I think I have to add in a
              place for spaces like sergey said.

              Sergey: I have been fittling with the last sequence by attempting to
              add spaces, but I still can't get it to work for some reason. There
              really is no specific requirements, I'm just trying to pull that name
              out.

              Thank you both for the explanations, it was very helpful, I'm just
              still having problems understanding why it won't work. The latest one
              I used was "([a-zA-Z_$][a-zA-Z0-9_$]\s*)"

              -Sean

                Comment

                • Sergey Zyuzin
                  #5
                  Re: Problem Creating Regex Expression

                  On Feb 8, 4:13 pm, Sean <ColdFusion...@ gmail.comwrote:
                  On Feb 8, 8:19 am, "Kevin Spencer" <unclechutney@l ocalhostwrote:
                  >
                  >
                  >
                  >
                  >
                  The following will work capture all content inside <td></tdtags:
                  >
                  (?<=td>)(.*?)(? =<)
                  >
                  The first part is a positive look-ahead, indicating that a match must be
                  preceded by the character sequence "td>" (non-capturing). The second part
                  indicates any character 0 or more times with a lazy quantifier, meaning that
                  it will capture as few times as possible. The third part is a positive
                  look-ahead, indicating that the match must be followed by a "<" character.
                  Since there are no "<" characters in the actual tag's content, this stops
                  the match at the end of the tag.
                  >
                  --
                  HTH,
                  >
                  Kevin Spencer
                  Chicken Salad Surgeon
                  Microsoft MVP
                  >
                  "Sean" <ColdFusion...@ gmail.comwrote in message
                  >
                  news:9e791c78-b1a2-43cd-916f-ae0202d31b66@1g 2000hsl.googleg roups.com...
                  >
                  >I am finally taking the time to get to know regex, but it seems I have
                  taken a bit of a tumble.
                  >
                  I have the following (dummy) data:
                  >
                  <td>Name:</td <td>Kherie Kali</td>
                  >
                  If I use this expression: "<td>Name:</td>\s*<td>Kheri e Kali</td>"
                  >
                  I indeed get a match.
                  >
                  The next step I took is to get this without knowing the name in
                  advance. I then used the following expression: <td>Name:</td>
                  \s*<td>([a-zA-Z_$][a-zA-Z0-9_$]*)</td>
                  >
                  and I didn't get a thing.
                  >
                  Shouldn't the "([a-zA-Z_$][a-zA-Z0-9_$]*)" part match any word or
                  symbol and then the * make it multiple words or symbols? Any light
                  that could be shed on this situation is appreciated. I don't
                  necessarily want the answer to my quandry, but insight as to what I am
                  doing wrong.
                  >
                  Thank you,
                  >
                  -Sean- Hide quoted text -
                  >
                  - Show quoted text -
                  >
                  Kevin and Sergey,
                  >
                  Kevin: Thank you for the explanation! This find the Name: Tag, but
                  won't find the dummy person's actual name. I think I have to add in a
                  place for spaces like sergey said.
                  >
                  Sergey: I have been fittling with the last sequence by attempting to
                  add spaces, but I still can't get it to work for some reason. There
                  really is no specific requirements, I'm just trying to pull that name
                  out.
                  >
                  Thank you both for the explanations, it was very helpful, I'm just
                  still having problems understanding why it won't work. The latest one
                  I used was "([a-zA-Z_$][a-zA-Z0-9_$]\s*)"
                  >
                  -Sean- Hide quoted text -
                  >
                  - Show quoted text -
                  You should put \s inside brackets "([a-zA-Z_$][a-zA-Z0-9_$\s]*)"
                  If you don't have specific requirements than probably you could use
                  expression
                  similar to what Kevin suggests or something like "<td>Name:</td>
                  \s*<td>(.*?)</td>"

                  Thanks,
                  Sergey

                    Comment

                    • Sean
                      #6
                      Re: Problem Creating Regex Expression

                      On Feb 8, 9:50 am, Sergey Zyuzin <forever....@gm ail.comwrote:
                      On Feb 8, 4:13 pm, Sean <ColdFusion...@ gmail.comwrote:
                      >
                      >
                      >
                      >
                      >
                      On Feb 8, 8:19 am, "Kevin Spencer" <unclechutney@l ocalhostwrote:
                      >
                      The following will work capture all content inside <td></tdtags:
                      >
                      (?<=td>)(.*?)(? =<)
                      >
                      The first part is a positive look-ahead, indicating that a match must be
                      preceded by the character sequence "td>" (non-capturing). The second part
                      indicates any character 0 or more times with a lazy quantifier, meaning that
                      it will capture as few times as possible. The third part is a positive
                      look-ahead, indicating that the match must be followed by a "<" character.
                      Since there are no "<" characters in the actual tag's content, this stops
                      the match at the end of the tag.
                      >
                      --
                      HTH,
                      >
                      Kevin Spencer
                      Chicken Salad Surgeon
                      Microsoft MVP
                      >
                      "Sean" <ColdFusion...@ gmail.comwrote in message
                      >
                      >news:9e791c7 8-b1a2-43cd-916f-ae0202d31b66@1g 2000hsl.googleg roups.com...
                      >
                      I am finally taking the time to get to know regex, but it seems I have
                      taken a bit of a tumble.
                      >
                      I have the following (dummy) data:
                      >
                      <td>Name:</td <td>Kherie Kali</td>
                      >
                      If I use this expression: "<td>Name:</td>\s*<td>Kheri e Kali</td>"
                      >
                      I indeed get a match.
                      >
                      The next step I took is to get this without knowing the name in
                      advance. I then used the following expression: <td>Name:</td>
                      \s*<td>([a-zA-Z_$][a-zA-Z0-9_$]*)</td>
                      >
                      and I didn't get a thing.
                      >
                      Shouldn't the "([a-zA-Z_$][a-zA-Z0-9_$]*)" part match any word or
                      symbol and then the * make it multiple words or symbols? Any light
                      that could be shed on this situation is appreciated. I don't
                      necessarily want the answer to my quandry, but insight as to what I am
                      doing wrong.
                      >
                      Thank you,
                      >
                      -Sean- Hide quoted text -
                      >
                      - Show quoted text -
                      >
                      Kevin and Sergey,
                      >
                      Kevin: Thank you for the explanation! This find the Name: Tag, but
                      won't find the dummy person's actual name. I think I have to add in a
                      place for spaces like sergey said.
                      >
                      Sergey: I have been fittling with the last sequence by attempting to
                      add spaces, but I still can't get it to work for some reason. There
                      really is no specific requirements, I'm just trying to pull that name
                      out.
                      >
                      Thank you both for the explanations, it was very helpful, I'm just
                      still having problems understanding why it won't work. The latest one
                      I used was "([a-zA-Z_$][a-zA-Z0-9_$]\s*)"
                      >
                      -Sean- Hide quoted text -
                      >
                      - Show quoted text -
                      >
                      You should put \s inside brackets "([a-zA-Z_$][a-zA-Z0-9_$\s]*)"
                      If you don't have specific requirements than probably you could use
                      expression
                      similar to what Kevin suggests or something like "<td>Name:</td>
                      \s*<td>(.*?)</td>"
                      >
                      Thanks,
                      Sergey- Hide quoted text -
                      >
                      - Show quoted text -
                      Perfect, thank you both!

                      I used your last suggestion Sergey, and did the following: "<td>Name:</
                      td>\s*<td>(?<a0 >(.*?))</td>" which correctly matched "Kherie Kali" and
                      put it into the a0 group.

                      I appreciate both of your help!

                      -Sean

                        Comment

                        Previous template Next
                        Working...