The flood of documents AC_ARG_ENABLE configure.in A section of code as follows : Dnl find libidn "= 0.3.0 AC_ARG_ENABLE (idn, AC_HELP_STRING ([--enable-idn], [enable a protocol support (yes)]), want_idn=$enableval, want_idn=yes) If test "x-$want_idn" = "x-yes" then AC_CHECK_HEADERS (stringprep.h) If test "x-$ac_cv_header_stringprep_h" = "x-yes" then AC_CHECK_LIB (idn, stringprep_check_version) Fi If test "x-$ac_cv_lib_idn_stringprep_check_version" = "x-yes" then AC_MSG_CHECKING (for Libidn version "= 0.3.0) AC_RUN_IFELSE ([AC_LANG_PROGRAM ([[#include "stringprep.h>]]. [[return! (Stringprep_check_version ( "0.3.0"))])] [AC_MSG_RESULT (Yes) Have_idn=yes]. AC_MSG_RESULT (no)) Fi If test "x-$have_idn" = "X" then AC_MSG_ERROR ([Libidn "not found] = 0.3.0) Fi AC_DEFINE (HAVE_IDN,1, if Libidn is available.] [Define to 1) Fi According to my understanding of the original input parameters : I --enbale-idn after AC_ARG_ENABLE want_idn=yes sentence will be implemented, and then if test "x-$want_idn" = "x-yes" judgments Now and then the implementation of the definition and implementation of AC_DEFINE (HAVE_IDN,1, if Libidn is available.] [Define to 1) Sentence HAVE_IDN macro definition. However, a look autoconf materials : AC_ARG_ENABLE (feature, help-string [action-if-given [action-if-not-given]]); I imported --enbale-idn so it will be implemented after AC_ARG_ENABLE want_idn=$enableval, according to this line of thinking, not the following sentence is not enforced, it will not be HAVE_IDN macro definition, but the fact is that the definition of the back, who helped me $enableval Seem to understand the value of their acquiescence is yes, here with the importation of --enable-idn is the same, I said that right? [ Closed window ]