Fossil SCM

Make it easier to exclude all robots by setting robot-restrict to "*" and still allowing pulls to occur and the login page to be accessed.

drh 2026-08-06 11:44 UTC trunk

Commit 35a09aea15efbe5134e95c7c07f896d752410360c883858c35a7cafd00990d11

Parent e81cac374b73f1c…

2 files changed +13 -1 +2 -1

M src/robot.c

+13 -1

		--- src/robot.c
		+++ src/robot.c
		@@ -422,11 +422,18 @@
422	422
423	423	/*
424	424	** Return true if one or more of the conditions below are true.
425	425	** Return false if all of the following are false:
426	426	**
427		-** * The zTag is on the robot-restrict list
	427	+** * The zTag is on the robot-restrict list and there exists
	428	+** some tags that do not match robot-restrict or else zTag is
	429	+** neither "/" or "xfer" or "login".
	430	+**
	431	+** The complication of the previous rule (everything after the
	432	+** first "and") prevents a robot-restrict setting that contains
	433	+** "*" from trying to bring up a robot-captcha for a "sync"
	434	+** operation (the "/" or "xfer" page) or for the "login" page.
428	435	**
429	436	** * The client that submitted the HTTP request might be
430	437	** a robot
431	438	**
432	439	** * The Request URI does not match any of the exceptions
		@@ -446,10 +453,15 @@
446	453	if( !robot_restrict_has_tag(zTag) ) return 0;
447	454	if( !client_might_be_a_robot() ) return 0;
448	455	if( robot_exception() ){
449	456	robot.resultCache = KNOWN_NOT_ROBOT;
450	457	return 0;
	458	+ }
	459	+ if( robot_restrict_has_tag("-not-a-real-tag-")
	460	+ && (strcmp(zTag,"login")==0 \|\| strcmp(zTag,"xfer")==0 \|\| zTag[0]=='/')
	461	+ ){
	462	+ return 0;
451	463	}
452	464	return 1;
453	465	}
454	466
455	467	/*
456	468

	--- src/robot.c
	+++ src/robot.c
	@@ -422,11 +422,18 @@
422
423	/*
424	** Return true if one or more of the conditions below are true.
425	** Return false if all of the following are false:
426	**
427	** * The zTag is on the robot-restrict list







428	**
429	** * The client that submitted the HTTP request might be
430	** a robot
431	**
432	** * The Request URI does not match any of the exceptions
	@@ -446,10 +453,15 @@
446	if( !robot_restrict_has_tag(zTag) ) return 0;
447	if( !client_might_be_a_robot() ) return 0;
448	if( robot_exception() ){
449	robot.resultCache = KNOWN_NOT_ROBOT;
450	return 0;





451	}
452	return 1;
453	}
454
455	/*
456

	--- src/robot.c
	+++ src/robot.c
	@@ -422,11 +422,18 @@
422
423	/*
424	** Return true if one or more of the conditions below are true.
425	** Return false if all of the following are false:
426	**
427	** * The zTag is on the robot-restrict list and there exists
428	** some tags that do not match robot-restrict or else zTag is
429	** neither "/" or "xfer" or "login".
430	**
431	** The complication of the previous rule (everything after the
432	** first "and") prevents a robot-restrict setting that contains
433	** "*" from trying to bring up a robot-captcha for a "sync"
434	** operation (the "/" or "xfer" page) or for the "login" page.
435	**
436	** * The client that submitted the HTTP request might be
437	** a robot
438	**
439	** * The Request URI does not match any of the exceptions
	@@ -446,10 +453,15 @@
453	if( !robot_restrict_has_tag(zTag) ) return 0;
454	if( !client_might_be_a_robot() ) return 0;
455	if( robot_exception() ){
456	robot.resultCache = KNOWN_NOT_ROBOT;
457	return 0;
458	}
459	if( robot_restrict_has_tag("-not-a-real-tag-")
460	&& (strcmp(zTag,"login")==0 \|\| strcmp(zTag,"xfer")==0 \|\| zTag[0]=='/')
461	){
462	return 0;
463	}
464	return 1;
465	}
466
467	/*
468

M src/setup.c

+2 -1

		--- src/setup.c
		+++ src/setup.c
		@@ -503,11 +503,12 @@
503	503	@ then it only applies if query parameters are such that
504	504	@ the page is expensive and/or unusual.
505	505	@ <li> The "ext" tag covers all extensions, but a tag like
506	506	@ "ext/PATH" only covers the specific extension at PATH.
507	507	@ </ul>
508		- @ To disable robot restrictions, change this setting to "off".
	508	+ @ Make this setting "off" to disable all robot restrictions, or "*"
	509	+ @ to exclude robots from all pages other than "/login" and "/xfer".
509	510	@ (Property: <a href="%R/help/robot-restrict">robot-restrict</a>)
510	511	@ <br>
511	512	textarea_attribute("", 2, 80,
512	513	"robot-restrict", "rbrestrict", robot_restrict_default(), 0);
513	514
514	515

	--- src/setup.c
	+++ src/setup.c
	@@ -503,11 +503,12 @@
503	@ then it only applies if query parameters are such that
504	@ the page is expensive and/or unusual.
505	@ <li> The "ext" tag covers all extensions, but a tag like
506	@ "ext/PATH" only covers the specific extension at PATH.
507	@ </ul>
508	@ To disable robot restrictions, change this setting to "off".

509	@ (Property: <a href="%R/help/robot-restrict">robot-restrict</a>)
510	@ <br>
511	textarea_attribute("", 2, 80,
512	"robot-restrict", "rbrestrict", robot_restrict_default(), 0);
513
514

	--- src/setup.c
	+++ src/setup.c
	@@ -503,11 +503,12 @@
503	@ then it only applies if query parameters are such that
504	@ the page is expensive and/or unusual.
505	@ <li> The "ext" tag covers all extensions, but a tag like
506	@ "ext/PATH" only covers the specific extension at PATH.
507	@ </ul>
508	@ Make this setting "off" to disable all robot restrictions, or "*"
509	@ to exclude robots from all pages other than "/login" and "/xfer".
510	@ (Property: <a href="%R/help/robot-restrict">robot-restrict</a>)
511	@ <br>
512	textarea_attribute("", 2, 80,
513	"robot-restrict", "rbrestrict", robot_restrict_default(), 0);
514
515