make_us_regexes 3.7 KB

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061
  1. #!/bin/sh
  2. ###########################################################################
  3. ## ##
  4. ## Language Technologies Institute ##
  5. ## Carnegie Mellon University ##
  6. ## Copyright (c) 2008 ##
  7. ## All Rights Reserved. ##
  8. ## ##
  9. ## Permission is hereby granted, free of charge, to use and distribute ##
  10. ## this software and its documentation without restriction, including ##
  11. ## without limitation the rights to use, copy, modify, merge, publish, ##
  12. ## distribute, sublicense, and/or sell copies of this work, and to ##
  13. ## permit persons to whom this work is furnished to do so, subject to ##
  14. ## the following conditions: ##
  15. ## 1. The code must retain the above copyright notice, this list of ##
  16. ## conditions and the following disclaimer. ##
  17. ## 2. Any modifications must be clearly marked as such. ##
  18. ## 3. Original authors' names are not deleted. ##
  19. ## 4. The authors' names are not used to endorse or promote products ##
  20. ## derived from this software without specific prior written ##
  21. ## permission. ##
  22. ## ##
  23. ## CARNEGIE MELLON UNIVERSITY AND THE CONTRIBUTORS TO THIS WORK ##
  24. ## DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ##
  25. ## ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT ##
  26. ## SHALL CARNEGIE MELLON UNIVERSITY NOR THE CONTRIBUTORS BE LIABLE ##
  27. ## FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES ##
  28. ## WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN ##
  29. ## AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ##
  30. ## ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF ##
  31. ## THIS SOFTWARE. ##
  32. ## ##
  33. ###########################################################################
  34. ## ##
  35. ## Make regexes into compiled form ##
  36. ## ##
  37. ###########################################################################
  38. COMPILE_REGEX=$1/compile_regexes
  39. echo "/* Autogenerated from make_us_regexes */"
  40. $COMPILE_REGEX ordinal_number "[0-9][0-9,]*\\(th\\|TH\\|st\\|ST\\|nd\\|ND\\|rd\\|RD\\)"
  41. $COMPILE_REGEX hasvowel ".*[aeiouAEIOU].*"
  42. $COMPILE_REGEX usmoney "\\\$[0-9,]+\\(\\.[0-9]+\\)?"
  43. $COMPILE_REGEX illion ".*illion"
  44. $COMPILE_REGEX romannums "\\(II?I?\\|IV\\|VI?I?I?\\|IX\\|X[VIX]*\\)"
  45. $COMPILE_REGEX drst "\\([dD][Rr]\\|[Ss][tT]\\)"
  46. $COMPILE_REGEX numess "[0-9]+s"
  47. $COMPILE_REGEX sevenphonenumber "[0-9][0-9][0-9]-[0-9][0-9][0-9][0-9]"
  48. $COMPILE_REGEX fourdigits "[0-9][0-9][0-9][0-9]"
  49. $COMPILE_REGEX threedigits "[0-9][0-9][0-9]"
  50. $COMPILE_REGEX numbertime "[0-9]?[0-9]:[0-5][0-9]"
  51. $COMPILE_REGEX numbertimexm "[0-9]?[0-9][:\\.][0-5][0-9][ap]m"
  52. $COMPILE_REGEX dottedabbrevs "\\([A-Za-z]\\.\\)+[A-Za-z]\\.?"
  53. $COMPILE_REGEX digitsslashdigits "[0-9]+/[0-9]+"
  54. $COMPILE_REGEX digits2dash "\\([0-9]+-.\\)+[0-9]+"
  55. $COMPILE_REGEX wandm "[0-9,]*[0-9]+\\(lb\\|LB\\|lbs\\|LBS\\|ft\\|FT\\|kg\\|km\\|oz\\|hz\\|Hz\\|HZ\\|KHz\\|MHz\\|GHz\\|cm\\|mm\\|ml\\|KB\\|MB\\|GB\\|TB\\)"
  56. exit 0