repo stringlengths 5 53 | pr_number int32 1 321k | task_type stringclasses 2
values | issue_text stringlengths 0 81.2k | pr_title stringlengths 1 319 | pr_body stringlengths 0 105k | base_sha stringlengths 40 40 | head_sha stringlengths 40 40 | gold_diff stringlengths 0 202M | changed_files listlengths 0 100 | review_threads listlengths 0 100 | test_patch stringlengths 0 23.4M | merged bool 1
class |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
ucb-bar/chipyard | 2,212 | comment_to_fix | Allow "lite" Chipyard builds with a minimal subset of submodules | For now, why not add this for every SBT project Chipyard depends on (for most this would automatically add the dependency). | 06f3a6215658405b706f0faf14ba6748f72ca607 | b69b3592e161165e0f92337f58a28611aa94dd90 | diff --git a/build.sbt b/build.sbt
index d9002e56fe..ae8b585720 100644
--- a/build.sbt
+++ b/build.sbt
@@ -153,20 +153,57 @@ lazy val testchipip = (project in file("generators/testchipip"))
.settings(libraryDependencies ++= rocketLibDeps.value)
.settings(commonSettings)
-lazy val chipyard = (project in file("ge... | [
"build.sbt"
] | [
{
"comment": "For now, why not add this for every SBT project Chipyard depends on (for most this would automatically add the dependency).",
"path": "build.sbt",
"hunk": "@@ -153,21 +153,34 @@ lazy val testchipip = (project in file(\"generators/testchipip\"))\n .settings(libraryDependencies ++= rocketL... | true | ||
ucb-bar/chipyard | 2,212 | comment_to_fix | Allow "lite" Chipyard builds with a minimal subset of submodules | here, can we pick one of the two? And then we should make this a convention that everyone agrees on. | 06f3a6215658405b706f0faf14ba6748f72ca607 | b69b3592e161165e0f92337f58a28611aa94dd90 | diff --git a/build.sbt b/build.sbt
index d9002e56fe..ae8b585720 100644
--- a/build.sbt
+++ b/build.sbt
@@ -153,20 +153,57 @@ lazy val testchipip = (project in file("generators/testchipip"))
.settings(libraryDependencies ++= rocketLibDeps.value)
.settings(commonSettings)
-lazy val chipyard = (project in file("ge... | [
"build.sbt"
] | [
{
"comment": "here, can we pick one of the two? And then we should make this a convention that everyone agrees on.",
"path": "build.sbt",
"hunk": "@@ -153,20 +153,58 @@ lazy val testchipip = (project in file(\"generators/testchipip\"))\n .settings(libraryDependencies ++= rocketLibDeps.value)\n .sett... | true | ||
ucb-bar/chipyard | 2,291 | issue_to_patch | fix: disabling the circt bump CI until we fix for compatibilty | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | d2717bbcf222324fd72f304d839d0c5835e94c8a | eb85768ae83d3bdcee793097c33fb0c8cf0533e1 | diff --git a/.github/workflows/update-circt.yml b/.github/workflows/update-circt.yml
deleted file mode 100644
index bdbf11f1d2..0000000000
--- a/.github/workflows/update-circt.yml
+++ /dev/null
@@ -1,25 +0,0 @@
-name: update-circt
-
-# update weekly
-on:
- schedule:
- - cron: 0 0 * * 1
-
-defaults:
- run:
- she... | [
".github/workflows/update-circt.yml",
".github/workflows/update-circt.yml.disabled"
] | [] | true | ||
ucb-bar/chipyard | 2,288 | issue_to_patch | Switch to modern python3 in insert-includes.py | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 484fc3bdffb80042e79b9d465ca716752662459c | e7726d9284e0ebd6088acde6e9e91f7e31449bea | diff --git a/scripts/insert-includes.py b/scripts/insert-includes.py
index 22dd8e9d93..5a3bdf285b 100755
--- a/scripts/insert-includes.py
+++ b/scripts/insert-includes.py
@@ -1,4 +1,4 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
"""
replaces a `include with the full include file.
| [
"scripts/insert-includes.py"
] | [] | true | ||
ucb-bar/chipyard | 2,286 | issue_to_patch | Fix support for non-gnu SED | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 01c50fbf4286b5a95b144caec489b3f392264dd1 | 92b3661c99ff55db7efc475bab0a484b6d5f4db8 | diff --git a/common.mk b/common.mk
index 7f292b6239..2a774cc8f0 100644
--- a/common.mk
+++ b/common.mk
@@ -1,4 +1,3 @@
-SED ?= sed
# Note: Individual rules that use RISCV or external tools perform local checks to avoid
# blocking unrelated targets. Use $(require_riscv) and $(call require_cmd,<tool>) inside recipes.... | [
"common.mk",
"variables.mk"
] | [] | true | ||
ucb-bar/chipyard | 2,285 | issue_to_patch | Find a modern bash on MacOS | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | d519070dbde87fb07bf95e61b23c35fca5b74805 | 2df0872fc8dc18df6b96e4e40cee93084b3d1624 | diff --git a/common.mk b/common.mk
index 09ef07f6b3..7f292b6239 100644
--- a/common.mk
+++ b/common.mk
@@ -1,9 +1,9 @@
-SHELL=/bin/bash
SED ?= sed
# Note: Individual rules that use RISCV or external tools perform local checks to avoid
# blocking unrelated targets. Use $(require_riscv) and $(call require_cmd,<tool>... | [
"common.mk",
"variables.mk"
] | [] | true | ||
ucb-bar/chipyard | 2,284 | issue_to_patch | Allow overriding firtool in PATH with FIRTOOL_BIN flag | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 24551ed5ef654c66d367235d5cb310a47f9452a0 | 753a5ec9423a4c4e95dc16f1859b56965a16b5a1 | diff --git a/common.mk b/common.mk
index 9cd6073dae..09ef07f6b3 100644
--- a/common.mk
+++ b/common.mk
@@ -195,9 +195,9 @@ else
endif
$(SFC_MFC_TARGETS) &: $(FIRRTL_FILE) $(FINAL_ANNO_FILE) $(MFC_LOWERING_OPTIONS)
- $(call require_cmd,firtool)
+ $(call require_cmd,$(FIRTOOL_BIN))
rm -rf $(GEN_COLLATERAL_DIR)
- (s... | [
"common.mk",
"variables.mk"
] | [] | true | ||
ucb-bar/chipyard | 2,281 | issue_to_patch | Add full-config-finder: finds eligible CONFIG targets | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 8f98d45f01251c25a0d697cd1bf6b100c162f2a8 | 1e1495660683c8ee7569a15d8163a9fa663dc697 | diff --git a/.github/workflows/chipyard-full-flow.yml b/.github/workflows/chipyard-full-flow.yml
index 78fa6a1a3e..3de358c984 100644
--- a/.github/workflows/chipyard-full-flow.yml
+++ b/.github/workflows/chipyard-full-flow.yml
@@ -106,6 +106,7 @@ jobs:
source env.sh
cd sims/verilator
ma... | [
".github/workflows/chipyard-full-flow.yml",
"common.mk",
"docs/Customization/Keys-Traits-Configs.rst",
"docs/Simulation/Software-RTL-Simulation.rst",
"generators/chipyard/src/main/scala/ConfigFinder.scala"
] | [] | true | ||
ucb-bar/chipyard | 2,282 | issue_to_patch | Only check for RISCV if running a make target that references it | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 8f98d45f01251c25a0d697cd1bf6b100c162f2a8 | ebc836cdad9e67f6d929c05f247e9c1ca929ac6c | diff --git a/common.mk b/common.mk
index 4ff0c477dc..54765d4de4 100644
--- a/common.mk
+++ b/common.mk
@@ -1,11 +1,21 @@
SHELL=/bin/bash
SED ?= sed
-ifndef RISCV
-$(error RISCV is unset. Did you source the Chipyard auto-generated env file (which activates the default conda environment)?)
-else
-$(info Running with ... | [
"common.mk",
"sims/vcs/Makefile",
"sims/verilator/Makefile",
"sims/xcelium/Makefile"
] | [] | true | ||
ucb-bar/chipyard | 2,283 | issue_to_patch | Fix CI | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | bc48eeb7bdf30b9c32756da56bf86392e1c0bce8 | 52f7de76f5db440841a88159bfd57a051cb583da | diff --git a/sims/verilator/Makefile b/sims/verilator/Makefile
index f950dc9a5a..4fd149fb18 100644
--- a/sims/verilator/Makefile
+++ b/sims/verilator/Makefile
@@ -1,11 +1,6 @@
#########################################################################################
# verilator makefile
##############################... | [
"sims/verilator/Makefile"
] | [] | true | ||
ucb-bar/chipyard | 2,279 | issue_to_patch | Make scripts/init-submodules.sh mac friendly | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 973f8732d4ffeb8cc231edf109e84eef114e3d6e | 6e9810aa5b23ebfe7664986832d281ab0af62266 | diff --git a/scripts/utils.sh b/scripts/utils.sh
index 633be1033e..b6a8357f1d 100755
--- a/scripts/utils.sh
+++ b/scripts/utils.sh
@@ -1,19 +1,63 @@
-#/usr/bin/env bash
+#!/usr/bin/env bash
#######################################
# Common setup. Init MacOS compatibility
# variables.
# Globals:
-# READLINK
+# ... | [
"scripts/utils.sh"
] | [] | true | ||
ucb-bar/chipyard | 2,280 | issue_to_patch | Relax checks for vcs/verilator/xcelium in PATH | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 973f8732d4ffeb8cc231edf109e84eef114e3d6e | f0db80b61241b94da9b897d855e05278f0befa7f | diff --git a/common.mk b/common.mk
index a42a05360e..4ff0c477dc 100644
--- a/common.mk
+++ b/common.mk
@@ -184,6 +184,11 @@ else
endif
$(SFC_MFC_TARGETS) &: $(FIRRTL_FILE) $(FINAL_ANNO_FILE) $(MFC_LOWERING_OPTIONS)
+ @# Ensure firtool is available before invoking CIRCT to emit Verilog
+ @command -v firtool >/dev/nu... | [
"common.mk",
"sims/vcs/Makefile",
"sims/verilator/Makefile",
"sims/xcelium/Makefile"
] | [] | true | ||
ucb-bar/chipyard | 2,262 | issue_to_patch | Bump testchipip for DPI function changes | testchipip PR: https://github.com/ucb-bar/testchipip/pull/260
This should fix Xcelium running on SoC configs which use `SimDRAM`.
<!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free t... | 1c628f7f4fac8e7208ff088073559222b83b91f3 | 0abb22ccb0c6cba6d4d90652b41d8387b0d4c596 | diff --git a/generators/testchipip b/generators/testchipip
index 6c2626575b..b2c2ab0469 160000
--- a/generators/testchipip
+++ b/generators/testchipip
@@ -1,1 +1,1 @@
-Subproject commit 6c2626575b7aea293210922b2df8aa94bb301409
+Subproject commit b2c2ab0469114baaa6b10ad5f1bbf2f42b464afd
| [
"generators/testchipip"
] | [] | true | ||
ucb-bar/chipyard | 2,226 | issue_to_patch | Error handling in init submodules | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | c627b28ee22a8a1e6b344fe46a80d414cf48e465 | 8040cca83fa05056d0c0fc370442ca8cfdc0b059 | diff --git a/scripts/init-submodules-no-riscv-tools-nolog.sh b/scripts/init-submodules-no-riscv-tools-nolog.sh
index 63238e10bd..560fa16a3c 100755
--- a/scripts/init-submodules-no-riscv-tools-nolog.sh
+++ b/scripts/init-submodules-no-riscv-tools-nolog.sh
@@ -11,6 +11,22 @@ source $RDIR/scripts/utils.sh
common_setup
... | [
"scripts/init-submodules-no-riscv-tools-nolog.sh"
] | [
{
"comment": "Since this function is only used in init-submods, I think it should be in that file, not utils.sh",
"path": "scripts/init-submodules-no-riscv-tools-nolog.sh",
"hunk": "@@ -11,6 +11,22 @@ source $RDIR/scripts/utils.sh\n \n common_setup\n \n+# Custom error handler function\n+error_handler() ... | true | ||
ucb-bar/chipyard | 2,248 | issue_to_patch | How to configure asynchronous reset
### Background Work
- [X] Yes, I searched the [mailing list](https://groups.google.com/forum/#!forum/chipyard)
- [X] Yes, I searched the [documentation](https://chipyard.readthedocs.io/)
### Feature Description
Hello, I have encountered some questions and would like to seek advic... | Add support for async reset | Users can now use the `WithAsyncClockGroups` config fragment to coerce certain clock groups to use an asynchronous reset. Note that users will likely also want to use the new `WithAsyncResetRocketSubsystem` config fragment, which ensures the Rocket Chip Debug Module also uses an async reset. Omitting this will lead to ... | 0c91ddb1186c1b8c13fdb54bce6fa6f8e1d5cbe7 | 3bdabbe6853bd240766de43e224668a46e1916d7 | diff --git a/generators/chipyard/src/main/scala/clocking/ClockGroupAsyncResetCoercer.scala b/generators/chipyard/src/main/scala/clocking/ClockGroupAsyncResetCoercer.scala
new file mode 100644
index 0000000000..3deaeb3428
--- /dev/null
+++ b/generators/chipyard/src/main/scala/clocking/ClockGroupAsyncResetCoercer.scala
@... | [
"generators/chipyard/src/main/scala/clocking/ClockGroupAsyncResetCoercer.scala",
"generators/chipyard/src/main/scala/clocking/HasChipyardPRCI.scala",
"generators/chipyard/src/main/scala/config/RocketConfigs.scala",
"generators/chipyard/src/main/scala/config/fragments/SubsystemFragments.scala",
"generators/t... | [] | true | |
ucb-bar/chipyard | 2,234 | issue_to_patch | ADD: branch predictor mode support for tacit | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | d22734ebb4aaa2b20a2f01c667dadd3929d2725a | e453942207e187b93bf35388612fb821a00462bd | diff --git a/build.sbt b/build.sbt
index 3a6c6b28c4..d9002e56fe 100644
--- a/build.sbt
+++ b/build.sbt
@@ -265,7 +265,7 @@ lazy val nvdla = (project in file("generators/nvdla"))
.settings(commonSettings)
lazy val tacit = (project in file("generators/tacit"))
- .dependsOn(rocketchip, shuttle)
+ .dependsOn(rocket... | [
"build.sbt",
"generators/chipyard/src/main/scala/DigitalTop.scala",
"generators/chipyard/src/main/scala/config/fragments/TileFragments.scala",
"generators/rocket-chip",
"generators/tacit",
"software/baremetal-ide",
"software/tacit_decoder"
] | [] | true | ||
ucb-bar/chipyard | 2,232 | issue_to_patch | Add option for externally clocked GCD peripheral example | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 0e2dd13f71e6df5508e898128d503124f653186c | 8a6a19a49b98251e13fc73aee713bc0c7b1c2e1a | diff --git a/generators/chipyard/src/main/scala/config/AbstractConfig.scala b/generators/chipyard/src/main/scala/config/AbstractConfig.scala
index 495cc9d3ea..490b8ebca9 100644
--- a/generators/chipyard/src/main/scala/config/AbstractConfig.scala
+++ b/generators/chipyard/src/main/scala/config/AbstractConfig.scala
@@ -6... | [
"generators/chipyard/src/main/scala/config/AbstractConfig.scala",
"generators/chipyard/src/main/scala/config/MMIOAcceleratorConfigs.scala",
"generators/chipyard/src/main/scala/example/GCD.scala",
"generators/chipyard/src/main/scala/iobinders/IOBinders.scala"
] | [
{
"comment": "Seems like `WithGCDBusy..` is not really accurately named anymore, since there is a optionally a clock as well now.\r\nMaybe `WithGCDIOPunchthrough`?",
"path": "generators/chipyard/src/main/scala/iobinders/IOBinders.scala",
"hunk": "@@ -560,12 +560,42 @@ class WithNMITiedOff extends Compos... | true | ||
ucb-bar/chipyard | 2,232 | comment_to_fix | Add option for externally clocked GCD peripheral example | Seems like `WithGCDBusy..` is not really accurately named anymore, since there is a optionally a clock as well now.
Maybe `WithGCDIOPunchthrough`? | 0e2dd13f71e6df5508e898128d503124f653186c | 8a6a19a49b98251e13fc73aee713bc0c7b1c2e1a | diff --git a/generators/chipyard/src/main/scala/iobinders/IOBinders.scala b/generators/chipyard/src/main/scala/iobinders/IOBinders.scala
index 1e34c3f8dc..a923247617 100644
--- a/generators/chipyard/src/main/scala/iobinders/IOBinders.scala
+++ b/generators/chipyard/src/main/scala/iobinders/IOBinders.scala
@@ -560,12 +5... | [
"generators/chipyard/src/main/scala/iobinders/IOBinders.scala"
] | [
{
"comment": "Seems like `WithGCDBusy..` is not really accurately named anymore, since there is a optionally a clock as well now.\r\nMaybe `WithGCDIOPunchthrough`?",
"path": "generators/chipyard/src/main/scala/iobinders/IOBinders.scala",
"hunk": "@@ -560,12 +560,42 @@ class WithNMITiedOff extends Compos... | true | ||
ucb-bar/chipyard | 2,230 | issue_to_patch | Bump spike | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 91593f760ddb917d569bd1e17d92f31e2f6bdcff | 310f6cdcebae2f99a78378be1fe3ba27f577159d | diff --git a/generators/gemmini b/generators/gemmini
index 3b14b89c7f..7a7e0e17d8 160000
--- a/generators/gemmini
+++ b/generators/gemmini
@@ -1,1 +1,1 @@
-Subproject commit 3b14b89c7f1d004fc4f03850706abdff76d007ce
+Subproject commit 7a7e0e17d80669d7d32561a3ffc9721e943333af
diff --git a/generators/testchipip b/generato... | [
"generators/gemmini",
"generators/testchipip",
"toolchains/riscv-tools/riscv-isa-sim",
"toolchains/riscv-tools/riscv-spike-devices"
] | [] | true | ||
ucb-bar/chipyard | 2,231 | issue_to_patch | Incorporate RocketChip JTAG + GDB debugging instructions, improve docs/conf.py for local development | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 91593f760ddb917d569bd1e17d92f31e2f6bdcff | 4da1cfc30f10cfddb7b2d112da52c5e811ed60ee | diff --git a/docs/Advanced-Concepts/Chip-Communication.rst b/docs/Advanced-Concepts/Chip-Communication.rst
index bc913f685d..b8b129ff35 100644
--- a/docs/Advanced-Concepts/Chip-Communication.rst
+++ b/docs/Advanced-Concepts/Chip-Communication.rst
@@ -129,35 +129,250 @@ The default Chipyard designs instantiate the DTM c... | [
"docs/Advanced-Concepts/Chip-Communication.rst",
"docs/conf.py"
] | [] | true | ||
ucb-bar/chipyard | 2,214 | issue_to_patch | Remove cake-pattern requirement for no-IO devices | This adds a API to rocket-chip that allow devices to inject themselves into the BaseSubsystem without requiring a `CanHavePeripheryX` trait. This only works for devices that don't generate top-level IO.
This makes it possible to make these devices optional as the chipyard-lite PR #2212
**Related PRs / Issues**:... | 975c07ed826e8fc91e8c47f950c22e4b29b043a1 | c491707c61b79694300487e64607b61d391ef114 | diff --git a/build.sbt b/build.sbt
index 414ee48258..3a6c6b28c4 100644
--- a/build.sbt
+++ b/build.sbt
@@ -194,7 +194,7 @@ lazy val constellation = (project in file("generators/constellation"))
.settings(commonSettings)
lazy val fft_generator = (project in file("generators/fft-generator"))
- .dependsOn(rocketchi... | [
"build.sbt",
"generators/chipyard/src/main/scala/DigitalTop.scala",
"generators/chipyard/src/main/scala/example/InitZero.scala",
"generators/chipyard/src/main/scala/example/dsptools/GenericFIR.scala",
"generators/chipyard/src/main/scala/example/dsptools/StreamingPassthrough.scala",
"generators/fft-generat... | [] | true | ||
ucb-bar/chipyard | 2,225 | issue_to_patch | Conda env setup check in build | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 30c6bd2a15222ab3c18448536ff5b7fd5039285f | 783534cd1e4a1231d92e20f5dafed718219cb05c | diff --git a/scripts/build-setup.sh b/scripts/build-setup.sh
index a7e44349a2..e56182aa83 100755
--- a/scripts/build-setup.sh
+++ b/scripts/build-setup.sh
@@ -166,6 +166,14 @@ if run_step "1"; then
# create conda-lock only environment to be used in this section.
# done with cloning base then installing conda ... | [
"scripts/build-setup.sh"
] | [] | true | ||
ucb-bar/chipyard | 2,221 | issue_to_patch | Bump vector components to fix whole-register vill | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 6605482f6a99fd578c724dba13ae56f14507a4c0 | e3fccb1fec7f2f5cdb80d552e96e259bc429baa8 | diff --git a/generators/ara b/generators/ara
index 1d1918ebb5..0c4fd8813f 160000
--- a/generators/ara
+++ b/generators/ara
@@ -1,1 +1,1 @@
-Subproject commit 1d1918ebb5b46da115753cd1d97c409ced8467b9
+Subproject commit 0c4fd8813fa33d6a35a8d9cb27aa48302919d17f
diff --git a/generators/rocket-chip b/generators/rocket-chip
... | [
"generators/ara",
"generators/rocket-chip",
"generators/saturn",
"generators/shuttle"
] | [] | true | ||
ucb-bar/chipyard | 2,191 | issue_to_patch | Add Chipyard port of Zephyr RTOS | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | e97acc9fc67ed30c4467232afd3bf0bb5638c40f | a3bd834e5e17b12ea7524874adff6d1303ab9ea8 | diff --git a/.github/scripts/build-zephyr.sh b/.github/scripts/build-zephyr.sh
new file mode 100755
index 0000000000..a5eab3a887
--- /dev/null
+++ b/.github/scripts/build-zephyr.sh
@@ -0,0 +1,24 @@
+#!/bin/bash
+
+# turn echo on and error on earliest command
+set -ex
+
+# get shared variables
+SCRIPT_DIR="$( cd "$( dir... | [
".github/scripts/build-zephyr.sh",
".github/scripts/check-commit.sh",
".github/scripts/run-tests.sh",
".github/workflows/chipyard-run-tests.yml",
".gitmodules",
"conda-reqs/chipyard-extended.yaml",
"conda-reqs/conda-lock-reqs/conda-requirements-riscv-tools-linux-64-lean.conda-lock.yml",
"conda-reqs/co... | [
{
"comment": "Can you add this to .github/scripts/check-commit.sh",
"path": ".gitmodules",
"hunk": "@@ -154,3 +154,6 @@\n [submodule \"generators/vexiiriscv\"]\n \tpath = generators/vexiiriscv\n \turl = https://github.com/ucb-bar/vexiiriscv-tile.git\n+[submodule \"software/zephyr\"]",
"resolving_sha... | true | ||
ucb-bar/chipyard | 2,191 | comment_to_fix | Add Chipyard port of Zephyr RTOS | Can you add this to .github/scripts/check-commit.sh | e97acc9fc67ed30c4467232afd3bf0bb5638c40f | a3bd834e5e17b12ea7524874adff6d1303ab9ea8 | diff --git a/.gitmodules b/.gitmodules
index e8eee4a8e4..3dad92dc1d 100644
--- a/.gitmodules
+++ b/.gitmodules
@@ -151,6 +151,9 @@
[submodule "generators/vexiiriscv"]
path = generators/vexiiriscv
url = https://github.com/ucb-bar/vexiiriscv-tile.git
+[submodule "software/zephyrproject/zephyr"]
+ path = software/zep... | [
".gitmodules"
] | [
{
"comment": "Can you add this to .github/scripts/check-commit.sh",
"path": ".gitmodules",
"hunk": "@@ -154,3 +154,6 @@\n [submodule \"generators/vexiiriscv\"]\n \tpath = generators/vexiiriscv\n \turl = https://github.com/ucb-bar/vexiiriscv-tile.git\n+[submodule \"software/zephyr\"]",
"resolving_sha... | true | ||
ucb-bar/chipyard | 2,191 | comment_to_fix | Add Chipyard port of Zephyr RTOS | If you update this yaml, you have to rerun the update-conda-lockfiles script. | e97acc9fc67ed30c4467232afd3bf0bb5638c40f | a3bd834e5e17b12ea7524874adff6d1303ab9ea8 | diff --git a/conda-reqs/chipyard-extended.yaml b/conda-reqs/chipyard-extended.yaml
index 99d4332e9d..c78705ab83 100644
--- a/conda-reqs/chipyard-extended.yaml
+++ b/conda-reqs/chipyard-extended.yaml
@@ -63,6 +63,14 @@ dependencies:
- graphviz
- expect
+ # zephyr deps
+ - ninja
+ - gperf
+ - ccac... | [
"conda-reqs/chipyard-extended.yaml"
] | [
{
"comment": "If you update this yaml, you have to rerun the update-conda-lockfiles script.",
"path": "conda-reqs/chipyard-extended.yaml",
"hunk": "@@ -63,6 +63,14 @@ dependencies:\n - graphviz\n - expect\n \n+ # zephyr deps\n+ - ninja\n+ - gperf\n+ - ccache\n+ - dtc\n+ - west\... | true | ||
ucb-bar/chipyard | 2,224 | issue_to_patch | Remove HasHierarchicalBusTopology mention in doc | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 30c6bd2a15222ab3c18448536ff5b7fd5039285f | 6d49004e59291e16a7110769d119d3a10baf8796 | diff --git a/docs/Advanced-Concepts/Top-Testharness.rst b/docs/Advanced-Concepts/Top-Testharness.rst
index 483814e39a..6ebb5164e2 100644
--- a/docs/Advanced-Concepts/Top-Testharness.rst
+++ b/docs/Advanced-Concepts/Top-Testharness.rst
@@ -57,10 +57,10 @@ System
``generators/chipyard/src/main/scala/System.scala`` com... | [
"docs/Advanced-Concepts/Top-Testharness.rst"
] | [] | true | ||
ucb-bar/chipyard | 2,222 | issue_to_patch | Remove no-dont-touch hack - this bug has been fixed in newer CIRCT | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | 6605482f6a99fd578c724dba13ae56f14507a4c0 | 98fe72247cdff137a6d61608f3a5adfbb98dee7d | diff --git a/generators/chipyard/src/main/scala/config/MMIOAcceleratorConfigs.scala b/generators/chipyard/src/main/scala/config/MMIOAcceleratorConfigs.scala
index 3c8ca45a90..85424af619 100644
--- a/generators/chipyard/src/main/scala/config/MMIOAcceleratorConfigs.scala
+++ b/generators/chipyard/src/main/scala/config/MM... | [
"generators/chipyard/src/main/scala/config/MMIOAcceleratorConfigs.scala",
"generators/chipyard/src/main/scala/config/RocketConfigs.scala",
"generators/chipyard/src/main/scala/config/TutorialConfigs.scala"
] | [] | true | ||
ucb-bar/chipyard | 2,219 | issue_to_patch | small build setup edits to prevent common failures | **Edit 1**
At some point the `build-setup.sh` line:
```
conda-lock install --conda $CONDA_EXE $CONDA_ENV_ARG $LOCKFILE &&
```
Got changed to:
```
conda-lock install --conda $(which conda) $CONDA_ENV_ARG $LOCKFILE &&
```
Which does not work in all cases (such as BWRC and EDA machines), due to this behavior of... | cecdd295960422fd27af64dac79aa28f0275e86d | 8282a31dbe5dac69c8d732bef8597f3d93540ebe | diff --git a/conda-reqs/chipyard-extended.yaml b/conda-reqs/chipyard-extended.yaml
index 99d4332e9d..728bc58dc1 100644
--- a/conda-reqs/chipyard-extended.yaml
+++ b/conda-reqs/chipyard-extended.yaml
@@ -69,7 +69,7 @@ dependencies:
# hammer packages
- sty
- - open_pdks.sky130a
+ # - open_pdks.sky130a
... | [
"conda-reqs/chipyard-extended.yaml",
"conda-reqs/conda-lock-reqs/conda-requirements-riscv-tools-linux-64.conda-lock.yml",
"scripts/build-setup.sh"
] | [] | true | ||
ucb-bar/chipyard | 2,217 | issue_to_patch | Update Spike.rst | EXTRA_SPIKE_FLAGS is not a thing. Author probably meant EXTRA_SIM_FLAGS, which works as expected.
| cecdd295960422fd27af64dac79aa28f0275e86d | 456eff1e7fadfaec9a3c7047c90b15ffa5ff4d34 | diff --git a/docs/Software/Spike.rst b/docs/Software/Spike.rst
index 10fa842965..630d5c184a 100644
--- a/docs/Software/Spike.rst
+++ b/docs/Software/Spike.rst
@@ -43,7 +43,7 @@ Spike-as-a-Tile can be configured with custom IPC, commit logging, and other beh
.. code-block:: shell
- make CONFIG=SpikeUltraFastCon... | [
"docs/Software/Spike.rst"
] | [] | true | ||
ucb-bar/chipyard | 2,215 | issue_to_patch | Bump rocket chip | Included PRs:
- https://github.com/chipsalliance/rocket-chip/pull/3728
- https://github.com/chipsalliance/rocket-chip/pull/3731
<!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to... | 47d0a46520be698ebb031d3533adb5c33124e039 | 310914b1801534768cc1f902cb1f588d0024fa32 | diff --git a/generators/rocket-chip b/generators/rocket-chip
index 337f2abf57..c9289f5a92 160000
--- a/generators/rocket-chip
+++ b/generators/rocket-chip
@@ -1,1 +1,1 @@
-Subproject commit 337f2abf57f38a16c7354704020587c98035d337
+Subproject commit c9289f5a92f34ccbb62170bc507e7670b4989325
| [
"generators/rocket-chip"
] | [] | true | ||
ucb-bar/chipyard | 2,213 | issue_to_patch | Add .gitignore to tests for in-tree "cmake ." builds | <!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<section> labels to indicate which section
of t... | e97acc9fc67ed30c4467232afd3bf0bb5638c40f | 6b364aad3af18e00387b822b5c33a3d70d783ca7 | [
"tests/.gitignore"
] | [] | diff --git a/tests/.gitignore b/tests/.gitignore
index 5cbdb5a956..d55ea94528 100644
--- a/tests/.gitignore
+++ b/tests/.gitignore
@@ -1,3 +1,8 @@
+CMakeCache.txt
+CMakeCache.txt
+CMakeFiles/
+Makefile
+cmake_install.cmake
*.o
*.riscv
*.dump
| true | ||
ucb-bar/chipyard | 2,209 | issue_to_patch | Update to newer rocket-chip | This removes the BootROM copying hack
<!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of the changelog:<sectio... | 4f97e92eb069594ba890e506285fc8c4c6e59011 | e1ce495449a33d19eb65d70e7578408114c01cbc | diff --git a/common.mk b/common.mk
index 827eed6a1d..a42a05360e 100644
--- a/common.mk
+++ b/common.mk
@@ -23,7 +23,7 @@ HELP_COMPILATION_VARIABLES += \
" MFC_BASE_LOWERING_OPTIONS = override lowering options to pass to the MLIR FIRRTL compiler" \
" ASPECTS = comma separated list of Chisel aspec... | [
"common.mk",
"fpga/src/main/scala/vc707/Configs.scala",
"fpga/src/main/scala/vcu118/Configs.scala",
"generators/chipyard/src/main/scala/DigitalTop.scala",
"generators/chipyard/src/main/scala/config/fragments/PeripheralFragments.scala",
"generators/chipyard/src/main/scala/iobinders/IOBinders.scala",
"gen... | [] | true | ||
ucb-bar/chipyard | 2,207 | issue_to_patch | Update to Chisel 6.7, use new annotation API | Bump submodules to remove deprecated `ChiselAnnotation` API, which is going away in Chisel 7. Submodule PRs:
- https://github.com/chipsalliance/rocket-chip/pull/3724
- https://github.com/chipsalliance/diplomacy/pull/19
- https://github.com/ucb-bar/constellation/pull/82
Chisel release notes:
- https://github.... | 4edb82f37ba77f9a02a191d2e2b5a1dab4fc0448 | 4020e82fc484ddded0536a49e830a25b360bfaec | diff --git a/build.sbt b/build.sbt
index d06d8780b2..414ee48258 100644
--- a/build.sbt
+++ b/build.sbt
@@ -1,8 +1,8 @@
import Tests._
-val chisel6Version = "6.5.0"
+val chisel6Version = "6.7.0"
val chiselTestVersion = "6.0.0"
-val scalaVersionFromChisel = "2.13.12"
+val scalaVersionFromChisel = "2.13.16"
val chi... | [
"build.sbt",
"generators/constellation",
"generators/diplomacy",
"generators/rocket-chip"
] | [] | true | ||
ucb-bar/chipyard | 2,206 | issue_to_patch | Remove deprecated rocket-chip annotation API | See https://github.com/chipsalliance/rocket-chip/pull/3723
<!--
First, please ensure that the title of your PR is sufficient to include in the next changelog.
Refer to https://github.com/ucb-bar/chipyard/releases for examples and feel free to ask reviewers for help.
Then, make sure to label your PR with one of ... | 424249b49f8d465329c57d5c87d52d87a3e3defd | 85588cbf006646e815c691f6bd277138c737cd67 | diff --git a/generators/boom b/generators/boom
index 7184be9db9..3d305f2ff5 160000
--- a/generators/boom
+++ b/generators/boom
@@ -1,1 +1,1 @@
-Subproject commit 7184be9db9d48bd01689cf9dd429a4ac32b21105
+Subproject commit 3d305f2ff5dab9371d9c95754888d4b49baea556
diff --git a/generators/chipyard/src/main/scala/example/T... | [
"generators/boom",
"generators/chipyard/src/main/scala/example/TutorialTile.scala",
"generators/cva6",
"generators/ibex",
"generators/radiance",
"generators/riscv-sodor",
"generators/rocket-chip",
"generators/shuttle"
] | [] | true | ||
ucb-bar/chipyard | 731 | issue_to_patch | Correceted minor typos | Corrected a typo.
**Related issue**: <!-- if applicable -->
<!-- choose one -->
**Type of change**: other enhancement
<!-- choose one -->
**Impact**: other
**Release Notes**
<!-- Text from here to the end of the body will be considered for inclusion in the release notes for the version containing this pu... | f387c4b99424e869235f927aebcb7dabc643a6f5 | d799fd9e145d4148bbfc1e01022f6558855fa180 | diff --git a/docs/TileLink-Diplomacy-Reference/Register-Router.rst b/docs/TileLink-Diplomacy-Reference/Register-Router.rst
index 6f5e1aa0ed..4068d5cd36 100644
--- a/docs/TileLink-Diplomacy-Reference/Register-Router.rst
+++ b/docs/TileLink-Diplomacy-Reference/Register-Router.rst
@@ -46,7 +46,7 @@ objects with an offset.... | [
"docs/TileLink-Diplomacy-Reference/Register-Router.rst"
] | [] | true | ||
ucb-bar/chipyard | 774 | issue_to_patch | Avoid permissions conflict on shared protocjar.webcache | The Java protoc plugin (part of FIRRTL) by default caches protoc in /tmp, and does not delete it afterwards.
This causes permissions failures when multiple users try to install Chipyard on the same machine, due to permissions failures over accessing the same cache in /tmp.
A solution is to place the tmpdir within t... | f22051ca84b87725e75527ed06dff16faf4828fc | 7ef40c8d79ac0db68c51089fef837de9ce944109 | diff --git a/.java_tmp/.gitignore b/.java_tmp/.gitignore
new file mode 100644
index 0000000000..5e7d2734cf
--- /dev/null
+++ b/.java_tmp/.gitignore
@@ -0,0 +1,4 @@
+# Ignore everything in this directory
+*
+# Except this file
+!.gitignore
diff --git a/variables.mk b/variables.mk
index 88ba73ee21..2843e53e54 100644
--- ... | [
".java_tmp/.gitignore",
"variables.mk"
] | [] | true | ||
ueberdosis/tiptap | 7,905 | issue_to_patch | Ordered lists do not support list types (`a`, `A`, `i` and `I`)
### What’s the bug you are facing?
When you create an ordered list with some `type` (alphabet order, roman numerals), these list types are disappearing when this kind of content is entered to Tiptap.
This is typical behavior for example when copying... | feat(extension-list): support ordered list type attribute (a, A, i, I) | ## Changes Overview
Ordered lists now support the HTML `type` attribute for alphabetical (`a`, `A`) and roman numeral (`i`, `I`) list markers. The type is preserved through HTML round-trips, paste from external editors (Google Docs, Word, LibreOffice), and markdown parsing/serialization.
## Implementation Approach
W... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/.changeset/2026-06-01-ordered-list-type-attribute.md b/.changeset/2026-06-01-ordered-list-type-attribute.md
new file mode 100644
index 0000000000..e4ff0aa95a
--- /dev/null
+++ b/.changeset/2026-06-01-ordered-list-type-attribute.md
@@ -0,0 +1,21 @@
+---
+'@tiptap/extension-list': minor
+'@tiptap/core': patc... | [
".changeset/2026-06-01-ordered-list-type-attribute.md",
"demos/src/Nodes/OrderedList/React/index.jsx",
"demos/src/Nodes/OrderedList/Vue/index.vue",
"demos/src/Nodes/OrderedList/index.spec.ts",
"packages/core/src/commands/toggleList.ts",
"packages/extension-list/__tests__/orderedListType.spec.ts",
"packa... | [
{
"comment": "`detectMarkerType` only treats \"a\"/\"A\" as alphabetical, so markers like \"b.\" or \"c)\" end up as the default numeric type. That breaks parsing/paste when lists start at a later alpha marker (should still map to type \"a\"/\"A\" and rely on `start` for the offset). Consider detecting roman fi... | diff --git a/demos/src/Nodes/OrderedList/index.spec.ts b/demos/src/Nodes/OrderedList/index.spec.ts
index f8b1a00e43..de494d2f10 100644
--- a/demos/src/Nodes/OrderedList/index.spec.ts
+++ b/demos/src/Nodes/OrderedList/index.spec.ts
@@ -82,6 +82,37 @@ test.describe(`${demoPath}/${demoName}`, () => {
await editor... | true |
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | `detectMarkerType` only treats "a"/"A" as alphabetical, so markers like "b." or "c)" end up as the default numeric type. That breaks parsing/paste when lists start at a later alpha marker (should still map to type "a"/"A" and rely on `start` for the offset). Consider detecting roman first, then mapping any all-lowercas... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/roman.ts b/packages/extension-list/src/ordered-list/roman.ts
new file mode 100644
index 0000000000..5c01414b31
--- /dev/null
+++ b/packages/extension-list/src/ordered-list/roman.ts
@@ -0,0 +1,295 @@
+const ROMAN_NUMERALS: [number, string][] = [
+ [1000, 'm'],
+ [9... | [
"packages/extension-list/src/ordered-list/roman.ts"
] | [
{
"comment": "`detectMarkerType` only treats \"a\"/\"A\" as alphabetical, so markers like \"b.\" or \"c)\" end up as the default numeric type. That breaks parsing/paste when lists start at a later alpha marker (should still map to type \"a\"/\"A\" and rely on `start` for the offset). Consider detecting roman fi... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | The plain-text `handlePaste` handler will run even when rich HTML is available on the clipboard. That can override the default HTML paste pipeline (and drop formatting / list nesting) whenever the plain-text happens to match the typed-marker regex. Consider bailing out when `text/html` exists so HTML paste (including `... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/ordered-list.ts b/packages/extension-list/src/ordered-list/ordered-list.ts
index 239b64817a..e4f891d72a 100644
--- a/packages/extension-list/src/ordered-list/ordered-list.ts
+++ b/packages/extension-list/src/ordered-list/ordered-list.ts
@@ -1,6 +1,15 @@
+import type... | [
"packages/extension-list/src/ordered-list/ordered-list.ts"
] | [
{
"comment": "The plain-text `handlePaste` handler will run even when rich HTML is available on the clipboard. That can override the default HTML paste pipeline (and drop formatting / list nesting) whenever the plain-text happens to match the typed-marker regex. Consider bailing out when `text/html` exists so H... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | The plain-text paste handler ignores the marker value for `start` (e.g. pasting `3. Item` produces a list starting at 1). It also doesn’t offset alpha/roman lists when the first marker is not the first item (e.g. `b. Item`). This loses list numbering on paste. | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/ordered-list.ts b/packages/extension-list/src/ordered-list/ordered-list.ts
index 239b64817a..e4f891d72a 100644
--- a/packages/extension-list/src/ordered-list/ordered-list.ts
+++ b/packages/extension-list/src/ordered-list/ordered-list.ts
@@ -1,6 +1,15 @@
+import type... | [
"packages/extension-list/src/ordered-list/ordered-list.ts"
] | [
{
"comment": "The plain-text paste handler ignores the marker value for `start` (e.g. pasting `3. Item` produces a list starting at 1). It also doesn’t offset alpha/roman lists when the first marker is not the first item (e.g. `b. Item`). This loses list numbering on paste.",
"path": "packages/extension-lis... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | `isBlockContentLine` only recognizes a very narrow set of typed ordered-list markers (`a`, roman letters, etc.). That means valid alpha markers like `b.`/`c)` won’t be treated as list-item boundaries, which can cause continuation lines to be grouped incorrectly during markdown tokenization. Reusing `ORDERED_LIST_ITEM_R... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/utils.ts b/packages/extension-list/src/ordered-list/utils.ts
index e50cddde95..bb5374ac61 100644
--- a/packages/extension-list/src/ordered-list/utils.ts
+++ b/packages/extension-list/src/ordered-list/utils.ts
@@ -5,12 +5,33 @@ import type {
MarkdownToken,
} from... | [
"packages/extension-list/src/ordered-list/utils.ts"
] | [
{
"comment": "`isBlockContentLine` only recognizes a very narrow set of typed ordered-list markers (`a`, roman letters, etc.). That means valid alpha markers like `b.`/`c)` won’t be treated as list-item boundaries, which can cause continuation lines to be grouped incorrectly during markdown tokenization. Reusin... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | For non-numeric markers, `itemNumber` is hard-coded to `1`. This makes the markdown tokenizer report `start: 1` even when the first marker is `3.`, `b.`, or `II.`, so numbering/type offsets are lost during parsing (and nested list `start` values become incorrect). You likely need to parse the marker into a numeric star... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/utils.ts b/packages/extension-list/src/ordered-list/utils.ts
index e50cddde95..bb5374ac61 100644
--- a/packages/extension-list/src/ordered-list/utils.ts
+++ b/packages/extension-list/src/ordered-list/utils.ts
@@ -5,12 +5,33 @@ import type {
MarkdownToken,
} from... | [
"packages/extension-list/src/ordered-list/utils.ts"
] | [
{
"comment": "For non-numeric markers, `itemNumber` is hard-coded to `1`. This makes the markdown tokenizer report `start: 1` even when the first marker is `3.`, `b.`, or `II.`, so numbering/type offsets are lost during parsing (and nested list `start` values become incorrect). You likely need to parse the mark... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | The markdown tokenizer start regex only accepts single-letter alpha markers (`[a-zA-Z]`), but `getListMarker` can serialize alpha lists beyond 26 items as multi-letter markers (e.g. `aa.`). That will break markdown round-trips for longer `type="a"/"A"` lists. Consider allowing `[a-zA-Z]+` (and updating the correspondin... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/ordered-list.ts b/packages/extension-list/src/ordered-list/ordered-list.ts
index 239b64817a..e4f891d72a 100644
--- a/packages/extension-list/src/ordered-list/ordered-list.ts
+++ b/packages/extension-list/src/ordered-list/ordered-list.ts
@@ -1,6 +1,15 @@
+import type... | [
"packages/extension-list/src/ordered-list/ordered-list.ts"
] | [
{
"comment": "The markdown tokenizer start regex only accepts single-letter alpha markers (`[a-zA-Z]`), but `getListMarker` can serialize alpha lists beyond 26 items as multi-letter markers (e.g. `aa.`). That will break markdown round-trips for longer `type=\"a\"/\"A\"` lists. Consider allowing `[a-zA-Z]+` (and... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | The plain-text paste tests validate line matching but don’t assert the key behavior introduced here: deriving `orderedList.attrs.type` and `orderedList.attrs.start` from the pasted marker (e.g. `3.`, `b.`, `II.`). As written, these tests wouldn’t catch regressions in the paste handler logic (and currently don’t reflect... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/__tests__/orderedListType.spec.ts b/packages/extension-list/__tests__/orderedListType.spec.ts
new file mode 100644
index 0000000000..ddf0c22fef
--- /dev/null
+++ b/packages/extension-list/__tests__/orderedListType.spec.ts
@@ -0,0 +1,877 @@
+import type { JSONContent } from '@tiptap/... | [
"packages/extension-list/__tests__/orderedListType.spec.ts"
] | [
{
"comment": "The plain-text paste tests validate line matching but don’t assert the key behavior introduced here: deriving `orderedList.attrs.type` and `orderedList.attrs.start` from the pasted marker (e.g. `3.`, `b.`, `II.`). As written, these tests wouldn’t catch regressions in the paste handler logic (and c... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | `ORDERED_LIST_MARKER_PATTERN` currently allows unlimited alphabetic markers (`[a-zA-Z]+`), but the alpha conversion logic only supports 1–2 letters. This can mis-detect normal text like `abc. Something` as an ordered list marker and then compute an incorrect start value (falls back to 1). Tightening the pattern to matc... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/utils.ts b/packages/extension-list/src/ordered-list/utils.ts
index e50cddde95..bb5374ac61 100644
--- a/packages/extension-list/src/ordered-list/utils.ts
+++ b/packages/extension-list/src/ordered-list/utils.ts
@@ -5,12 +5,33 @@ import type {
MarkdownToken,
} from... | [
"packages/extension-list/src/ordered-list/utils.ts"
] | [
{
"comment": "`ORDERED_LIST_MARKER_PATTERN` currently allows unlimited alphabetic markers (`[a-zA-Z]+`), but the alpha conversion logic only supports 1–2 letters. This can mis-detect normal text like `abc. Something` as an ordered list marker and then compute an incorrect start value (falls back to 1). Tighteni... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | `parsePlainTextOrderedListPaste` builds the list attrs from only the first marker and doesn’t verify the rest of the pasted markers are consistent/sequential. That can convert non-list text (or non-sequential markers) into an ordered list whose generated markers don’t match what the user pasted. Add a quick validation ... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/utils.ts b/packages/extension-list/src/ordered-list/utils.ts
index e50cddde95..bb5374ac61 100644
--- a/packages/extension-list/src/ordered-list/utils.ts
+++ b/packages/extension-list/src/ordered-list/utils.ts
@@ -5,12 +5,33 @@ import type {
MarkdownToken,
} from... | [
"packages/extension-list/src/ordered-list/utils.ts"
] | [
{
"comment": "`parsePlainTextOrderedListPaste` builds the list attrs from only the first marker and doesn’t verify the rest of the pasted markers are consistent/sequential. That can convert non-list text (or non-sequential markers) into an ordered list whose generated markers don’t match what the user pasted. A... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | The paste handler calls `schema.nodeFromJSON(...)` without any error handling. If a consumer’s schema can’t represent the generated content (e.g. listItem doesn’t allow paragraph, or paragraph is not present), this will throw during paste and can break the editor interaction. Wrap node creation/transaction in a try/cat... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/ordered-list.ts b/packages/extension-list/src/ordered-list/ordered-list.ts
index 239b64817a..e4f891d72a 100644
--- a/packages/extension-list/src/ordered-list/ordered-list.ts
+++ b/packages/extension-list/src/ordered-list/ordered-list.ts
@@ -1,6 +1,15 @@
+import type... | [
"packages/extension-list/src/ordered-list/ordered-list.ts"
] | [
{
"comment": "The paste handler calls `schema.nodeFromJSON(...)` without any error handling. If a consumer’s schema can’t represent the generated content (e.g. listItem doesn’t allow paragraph, or paragraph is not present), this will throw during paste and can break the editor interaction. Wrap node creation/tr... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | `parsePlainTextOrderedListPaste` filters out empty lines, which can silently drop structure (e.g. blank lines between items) and still convert the paste into a list. Building `lines` from `text.trim()` without filtering keeps the common trailing-newline case working while ensuring internal blank lines prevent list conv... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/utils.ts b/packages/extension-list/src/ordered-list/utils.ts
index e50cddde95..bb5374ac61 100644
--- a/packages/extension-list/src/ordered-list/utils.ts
+++ b/packages/extension-list/src/ordered-list/utils.ts
@@ -5,12 +5,33 @@ import type {
MarkdownToken,
} from... | [
"packages/extension-list/src/ordered-list/utils.ts"
] | [
{
"comment": "`parsePlainTextOrderedListPaste` filters out empty lines, which can silently drop structure (e.g. blank lines between items) and still convert the paste into a list. Building `lines` from `text.trim()` without filtering keeps the common trailing-newline case working while ensuring internal blank l... | true | ||
ueberdosis/tiptap | 7,905 | comment_to_fix | feat(extension-list): support ordered list type attribute (a, A, i, I) | `detectMarkerType` treats any-length alpha strings as list markers (`/^[a-z]+$/` / `/^[A-Z]+$/`), but the alpha start conversion only supports 1–2 letters (`fromAlpha`). This means inputs like `abc` are classified as alpha but can’t be converted consistently. Restrict alpha detection to the supported 1–2 letter range (... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | a2042a2d33cc17b1798523ec09c6ef1351d1e607 | diff --git a/packages/extension-list/src/ordered-list/roman.ts b/packages/extension-list/src/ordered-list/roman.ts
new file mode 100644
index 0000000000..5c01414b31
--- /dev/null
+++ b/packages/extension-list/src/ordered-list/roman.ts
@@ -0,0 +1,295 @@
+const ROMAN_NUMERALS: [number, string][] = [
+ [1000, 'm'],
+ [9... | [
"packages/extension-list/src/ordered-list/roman.ts"
] | [
{
"comment": "`detectMarkerType` treats any-length alpha strings as list markers (`/^[a-z]+$/` / `/^[A-Z]+$/`), but the alpha start conversion only supports 1–2 letters (`fromAlpha`). This means inputs like `abc` are classified as alpha but can’t be converted consistently. Restrict alpha detection to the suppor... | true | ||
ueberdosis/tiptap | 7,922 | issue_to_patch | Remove circular dependencies by importing from specific modules | ## Changes Overview
Refactored imports in the core package to eliminate circular dependencies by importing directly from specific modules instead of the barrel index file. Additionally moved the `Commands` and `Storage` type declarations from `index.ts` to `types.ts`.
## Implementation Approach
Changed imports in `E... | dec826412c73eee20131d73986916f7e420c1aaf | a28ad278fa3f48c444fc177e84dd4789fd53ff16 | diff --git a/packages/core/src/ExtensionManager.ts b/packages/core/src/ExtensionManager.ts
index 0ca4791e61..47496275c7 100644
--- a/packages/core/src/ExtensionManager.ts
+++ b/packages/core/src/ExtensionManager.ts
@@ -8,6 +8,7 @@ import {
flattenExtensions,
getAttributesFromExtensions,
getExtensionField,
+ g... | [
"packages/core/src/ExtensionManager.ts",
"packages/core/src/index.ts",
"packages/core/src/types.ts"
] | [
{
"comment": "The `oxlint-disable-next-line` directives here are very broad and will suppress any lint rule for the next line. In this file you already use the more targeted `oxlint-disable-next-lineno-empty-object-type` for empty interfaces, so it’d be better to use the same rule-specific disable for `Commands... | true | ||
ueberdosis/tiptap | 7,922 | comment_to_fix | Remove circular dependencies by importing from specific modules | The `oxlint-disable-next-line` directives here are very broad and will suppress any lint rule for the next line. In this file you already use the more targeted `oxlint-disable-next-lineno-empty-object-type` for empty interfaces, so it’d be better to use the same rule-specific disable for `Commands` and `Storage` to avo... | dec826412c73eee20131d73986916f7e420c1aaf | a28ad278fa3f48c444fc177e84dd4789fd53ff16 | diff --git a/packages/core/src/types.ts b/packages/core/src/types.ts
index 36904f30af..0fdbb9f8a8 100644
--- a/packages/core/src/types.ts
+++ b/packages/core/src/types.ts
@@ -20,16 +20,10 @@ import type {
import type { Editor } from './Editor.js'
import type { Extendable } from './Extendable.js'
-import type {
- C... | [
"packages/core/src/types.ts"
] | [
{
"comment": "The `oxlint-disable-next-line` directives here are very broad and will suppress any lint rule for the next line. In this file you already use the more targeted `oxlint-disable-next-lineno-empty-object-type` for empty interfaces, so it’d be better to use the same rule-specific disable for `Commands... | true | ||
ueberdosis/tiptap | 7,928 | issue_to_patch | Custom markdown tokenizers dropped when a dedicated `marked` instance is injected into MarkdownManager
### Affected package
`@tiptap/markdown` (≥ 3.21.0, current `main` included; verified on 3.26.0)
### Description
When a dedicated `marked` instance is **injected** into `MarkdownManager` (the recommended way to iso... | fix(markdown): run custom tokenizers on injected Marked instances | ## Changes Overview
Fixes #7927.
Custom markdown tokenizers were silently dropped whenever a dedicated `marked` instance was injected into `MarkdownManager` (a common pattern for isolating tokenizers across editors with different settings).
`MarkdownManager.parse()` builds its lexer through `createLexer()`, which ca... | 6c1efd373bafe7db2ebe551e3be8b3ca4f774a5d | 83309f87c62d1adde22d22d75a3faf571f8aef3e | diff --git a/.changeset/markdown-injected-marked-tokenizers.md b/.changeset/markdown-injected-marked-tokenizers.md
new file mode 100644
index 0000000000..90eaa80fd3
--- /dev/null
+++ b/.changeset/markdown-injected-marked-tokenizers.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/markdown': patch
+---
+
+Fix parsing with injected Mar... | [
".changeset/markdown-injected-marked-tokenizers.md",
"packages/markdown/__tests__/manager.spec.ts",
"packages/markdown/src/MarkdownManager.ts"
] | [
{
"comment": "This comment is super verbose, please minimize it a bit. one or two lines should be more than enough.",
"path": "packages/markdown/__tests__/manager.spec.ts",
"hunk": "@@ -259,6 +259,38 @@ Second paragraph.`\n },\n ])\n })\n+\n+ it('runs custom tokenizers registered on... | diff --git a/packages/markdown/__tests__/manager.spec.ts b/packages/markdown/__tests__/manager.spec.ts
index 35a94521b4..ffba29f024 100644
--- a/packages/markdown/__tests__/manager.spec.ts
+++ b/packages/markdown/__tests__/manager.spec.ts
@@ -259,6 +259,34 @@ Second paragraph.`
},
])
})
+
+ it('... | true |
ueberdosis/tiptap | 7,928 | comment_to_fix | fix(markdown): run custom tokenizers on injected Marked instances | This comment is super verbose, please minimize it a bit. one or two lines should be more than enough. | 6c1efd373bafe7db2ebe551e3be8b3ca4f774a5d | 83309f87c62d1adde22d22d75a3faf571f8aef3e | diff --git a/packages/markdown/__tests__/manager.spec.ts b/packages/markdown/__tests__/manager.spec.ts
index 35a94521b4..ffba29f024 100644
--- a/packages/markdown/__tests__/manager.spec.ts
+++ b/packages/markdown/__tests__/manager.spec.ts
@@ -259,6 +259,34 @@ Second paragraph.`
},
])
})
+
+ it('... | [
"packages/markdown/__tests__/manager.spec.ts"
] | [
{
"comment": "This comment is super verbose, please minimize it a bit. one or two lines should be more than enough.",
"path": "packages/markdown/__tests__/manager.spec.ts",
"hunk": "@@ -259,6 +259,38 @@ Second paragraph.`\n },\n ])\n })\n+\n+ it('runs custom tokenizers registered on... | true | ||
ueberdosis/tiptap | 7,928 | comment_to_fix | fix(markdown): run custom tokenizers on injected Marked instances | Please simplify the changeset. It's a bit long for a normal changelog. Users usually want to parse it quickly and understand what changes for them. | 6c1efd373bafe7db2ebe551e3be8b3ca4f774a5d | 83309f87c62d1adde22d22d75a3faf571f8aef3e | diff --git a/.changeset/markdown-injected-marked-tokenizers.md b/.changeset/markdown-injected-marked-tokenizers.md
new file mode 100644
index 0000000000..90eaa80fd3
--- /dev/null
+++ b/.changeset/markdown-injected-marked-tokenizers.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/markdown': patch
+---
+
+Fix parsing with injected Mar... | [
".changeset/markdown-injected-marked-tokenizers.md"
] | [
{
"comment": "Please simplify the changeset. It's a bit long for a normal changelog. Users usually want to parse it quickly and understand what changes for them.",
"path": ".changeset/markdown-injected-marked-tokenizers.md",
"hunk": "@@ -0,0 +1,5 @@\n+---\n+'@tiptap/markdown': patch\n+---\n+\n+Fix custo... | true | ||
ueberdosis/tiptap | 7,928 | comment_to_fix | fix(markdown): run custom tokenizers on injected Marked instances | Same here, please minimize. 5 lines of comments for one line of code is too much. | 6c1efd373bafe7db2ebe551e3be8b3ca4f774a5d | 83309f87c62d1adde22d22d75a3faf571f8aef3e | diff --git a/packages/markdown/src/MarkdownManager.ts b/packages/markdown/src/MarkdownManager.ts
index 0d1c9fd5c8..761818f36f 100644
--- a/packages/markdown/src/MarkdownManager.ts
+++ b/packages/markdown/src/MarkdownManager.ts
@@ -205,7 +205,8 @@ export class MarkdownManager {
}
private createLexer(): Lexer {
-... | [
"packages/markdown/src/MarkdownManager.ts"
] | [
{
"comment": "Same here, please minimize. 5 lines of comments for one line of code is too much.",
"path": "packages/markdown/src/MarkdownManager.ts",
"hunk": "@@ -205,7 +205,12 @@ export class MarkdownManager {\n }\n \n private createLexer(): Lexer {\n- return new this.markedInstance.Lexer()\n+ ... | true | ||
ueberdosis/tiptap | 7,928 | comment_to_fix | fix(markdown): run custom tokenizers on injected Marked instances | The changeset entry is very detailed (implementation/root-cause level). Changesets are used for user-facing changelogs, so this should be shortened to a concise description of the observable fix (custom tokenizers on injected Marked instances now work). | 6c1efd373bafe7db2ebe551e3be8b3ca4f774a5d | 83309f87c62d1adde22d22d75a3faf571f8aef3e | diff --git a/.changeset/markdown-injected-marked-tokenizers.md b/.changeset/markdown-injected-marked-tokenizers.md
new file mode 100644
index 0000000000..90eaa80fd3
--- /dev/null
+++ b/.changeset/markdown-injected-marked-tokenizers.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/markdown': patch
+---
+
+Fix parsing with injected Mar... | [
".changeset/markdown-injected-marked-tokenizers.md"
] | [
{
"comment": "The changeset entry is very detailed (implementation/root-cause level). Changesets are used for user-facing changelogs, so this should be shortened to a concise description of the observable fix (custom tokenizers on injected Marked instances now work).",
"path": ".changeset/markdown-injected-... | true | ||
ueberdosis/tiptap | 7,934 | issue_to_patch | Publish a new stable version | This PR was opened by the [Changesets release](https://github.com/changesets/action) GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will ... | a38c9c040c7f516f16e9306724edb9c4d62f23fe | d817854c8f741e867217e5384ea5c43e9f9e8dff | diff --git a/.changeset/2026-06-10-node-range-selection-json-id.md b/.changeset/2026-06-10-node-range-selection-json-id.md
deleted file mode 100644
index 7b87809fcd..0000000000
--- a/.changeset/2026-06-10-node-range-selection-json-id.md
+++ /dev/null
@@ -1,5 +0,0 @@
----
-"@tiptap/extension-node-range": patch
----
-
-F... | [
".changeset/2026-06-10-node-range-selection-json-id.md",
".changeset/2026-06-10-restore-node-range-after-drop.md",
".changeset/gorgeous-terms-sell.md",
"CHANGELOG.md",
"packages-deprecated/extension-character-count/CHANGELOG.md",
"packages-deprecated/extension-character-count/package.json",
"packages-de... | [] | true | ||
ueberdosis/tiptap | 7,930 | issue_to_patch | fix: resolve drag-and-drop duplication issue during collaboration | ## Changes Overview
Fixes drag-and-drop duplicating blocks during collaboration. When a remote collaborator edited the document while a drag was in progress, dropping left an empty copy of the dragged block at its original position.
The root cause is that `NodeRangeSelection` never registered a JSON id, so `selec... | e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/.changeset/2026-06-10-node-range-selection-json-id.md b/.changeset/2026-06-10-node-range-selection-json-id.md
new file mode 100644
index 0000000000..7b87809fcd
--- /dev/null
+++ b/.changeset/2026-06-10-node-range-selection-json-id.md
@@ -0,0 +1,5 @@
+---
+"@tiptap/extension-node-range": patch
+---
+
+Fixed... | [
".changeset/2026-06-10-node-range-selection-json-id.md",
".changeset/2026-06-10-restore-node-range-after-drop.md",
".changeset/gorgeous-terms-sell.md",
"demos/package.json",
"packages/extension-collaboration-caret/package.json",
"packages/extension-collaboration/package.json",
"packages/extension-drag-h... | [
{
"comment": "`depth` is now serialized/restored via `toJSON`/`fromJSON`, but `map()` still recreates a `NodeRangeSelection` without passing `this.depth`. This causes the selection’s depth to be dropped as soon as the selection is mapped through a transaction (including remote collaboration updates), underminin... | diff --git a/packages/extension-drag-handle/__tests__/nodeRangeDrop.spec.ts b/packages/extension-drag-handle/__tests__/nodeRangeDrop.spec.ts
new file mode 100644
index 0000000000..bac0f41628
--- /dev/null
+++ b/packages/extension-drag-handle/__tests__/nodeRangeDrop.spec.ts
@@ -0,0 +1,105 @@
+import { Editor } from '@ti... | true | |
ueberdosis/tiptap | 7,930 | comment_to_fix | fix: resolve drag-and-drop duplication issue during collaboration | `depth` is now serialized/restored via `toJSON`/`fromJSON`, but `map()` still recreates a `NodeRangeSelection` without passing `this.depth`. This causes the selection’s depth to be dropped as soon as the selection is mapped through a transaction (including remote collaboration updates), undermining the goal of lossless... | e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/packages/extension-node-range/src/helpers/NodeRangeSelection.ts b/packages/extension-node-range/src/helpers/NodeRangeSelection.ts
index e6a71a4094..ce6c5b08e4 100644
--- a/packages/extension-node-range/src/helpers/NodeRangeSelection.ts
+++ b/packages/extension-node-range/src/helpers/NodeRangeSelection.ts
@... | [
"packages/extension-node-range/src/helpers/NodeRangeSelection.ts"
] | [
{
"comment": "`depth` is now serialized/restored via `toJSON`/`fromJSON`, but `map()` still recreates a `NodeRangeSelection` without passing `this.depth`. This causes the selection’s depth to be dropped as soon as the selection is mapped through a transaction (including remote collaboration updates), underminin... | true | ||
ueberdosis/tiptap | 7,930 | comment_to_fix | fix: resolve drag-and-drop duplication issue during collaboration | This comment doesn’t match what the positions actually do: `head` is computed as `firstParagraphSize + secondParagraphSize`, which resolves to the start of the third paragraph (the boundary between the 2nd and 3rd blocks), not “inside the second paragraph”. Updating the comment will make the test intent clearer and avo... | e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/packages/extension-node-range/__tests__/NodeRangeSelection.spec.ts b/packages/extension-node-range/__tests__/NodeRangeSelection.spec.ts
new file mode 100644
index 0000000000..82cf9ead45
--- /dev/null
+++ b/packages/extension-node-range/__tests__/NodeRangeSelection.spec.ts
@@ -0,0 +1,93 @@
+import { Editor ... | [
"packages/extension-node-range/__tests__/NodeRangeSelection.spec.ts"
] | [
{
"comment": "This comment doesn’t match what the positions actually do: `head` is computed as `firstParagraphSize + secondParagraphSize`, which resolves to the start of the third paragraph (the boundary between the 2nd and 3rd blocks), not “inside the second paragraph”. Updating the comment will make the test ... | true | ||
ueberdosis/tiptap | 7,930 | comment_to_fix | fix: resolve drag-and-drop duplication issue during collaboration | The phrase “latest version” is time-dependent and will become inaccurate over time in the changelog. Prefer wording that describes the required capability (or a minimum version, if known) so readers can tell what they need to upgrade to. | e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/.changeset/2026-06-10-node-range-selection-json-id.md b/.changeset/2026-06-10-node-range-selection-json-id.md
new file mode 100644
index 0000000000..7b87809fcd
--- /dev/null
+++ b/.changeset/2026-06-10-node-range-selection-json-id.md
@@ -0,0 +1,5 @@
+---
+"@tiptap/extension-node-range": patch
+---
+
+Fixed... | [
".changeset/2026-06-10-node-range-selection-json-id.md"
] | [
{
"comment": "The phrase “latest version” is time-dependent and will become inaccurate over time in the changelog. Prefer wording that describes the required capability (or a minimum version, if known) so readers can tell what they need to upgrade to.",
"path": ".changeset/2026-06-10-node-range-selection-js... | true | ||
ueberdosis/tiptap | 7,930 | comment_to_fix | fix: resolve drag-and-drop duplication issue during collaboration | `NodeRangeBookmark.resolve()` still constructs new `NodeRangeSelection($anchor, $head)` without depth. map() is fixed, but bookmark-based restoration (local transactions that preserve selection across steps) can still lose depth. Lower risk for drag-handle (depth 0), but inconsistent with the depth work elsewhere.
| e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/packages/extension-node-range/src/helpers/NodeRangeSelection.ts b/packages/extension-node-range/src/helpers/NodeRangeSelection.ts
index e6a71a4094..ce6c5b08e4 100644
--- a/packages/extension-node-range/src/helpers/NodeRangeSelection.ts
+++ b/packages/extension-node-range/src/helpers/NodeRangeSelection.ts
@... | [
"packages/extension-node-range/src/helpers/NodeRangeSelection.ts"
] | [
{
"comment": "`NodeRangeBookmark.resolve()` still constructs new `NodeRangeSelection($anchor, $head)` without depth. map() is fixed, but bookmark-based restoration (local transactions that preserve selection across steps) can still lose depth. Lower risk for drag-handle (depth 0), but inconsistent with the dept... | true | ||
ueberdosis/tiptap | 7,930 | comment_to_fix | fix: resolve drag-and-drop duplication issue during collaboration | Race between drop and requestAnimationFrame restore
Restore runs one frame later and uses current selection.from. If the user clicks elsewhere, a plugin changes selection, or a remote Yjs update arrives in that frame, restore can select the wrong blocks or fail silently.
Suggestion: Capture `selection.from` (or the m... | e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/packages/extension-drag-handle/src/drag-handle-plugin.ts b/packages/extension-drag-handle/src/drag-handle-plugin.ts
index 6731a46947..7da9325e62 100644
--- a/packages/extension-drag-handle/src/drag-handle-plugin.ts
+++ b/packages/extension-drag-handle/src/drag-handle-plugin.ts
@@ -2,7 +2,7 @@ import { type... | [
"packages/extension-drag-handle/src/drag-handle-plugin.ts"
] | [
{
"comment": "Race between drop and requestAnimationFrame restore\n\nRestore runs one frame later and uses current selection.from. If the user clicks elsewhere, a plugin changes selection, or a remote Yjs update arrives in that frame, restore can select the wrong blocks or fail silently.\n\nSuggestion: Capture ... | true | ||
ueberdosis/tiptap | 7,930 | comment_to_fix | fix: resolve drag-and-drop duplication issue during collaboration | Document-level drop listener is not editor-scoped
`onDrop` is on document, not the editor DOM. Any drop on the page while activeDragRange is set schedules a restore. That pattern existed before, but the new restore dispatch makes side effects more visible — especially with multiple editors on one page.
Suggestion: Ch... | e45ea22e55cf5944513449f7a18a75c2cf68ee31 | cab7b91d1d52398a0ee9c59e03c4c406deb11f24 | diff --git a/packages/extension-drag-handle/src/drag-handle-plugin.ts b/packages/extension-drag-handle/src/drag-handle-plugin.ts
index 6731a46947..7da9325e62 100644
--- a/packages/extension-drag-handle/src/drag-handle-plugin.ts
+++ b/packages/extension-drag-handle/src/drag-handle-plugin.ts
@@ -2,7 +2,7 @@ import { type... | [
"packages/extension-drag-handle/src/drag-handle-plugin.ts"
] | [
{
"comment": "Document-level drop listener is not editor-scoped\n\n`onDrop` is on document, not the editor DOM. Any drop on the page while activeDragRange is set schedules a restore. That pattern existed before, but the new restore dispatch makes side effects more visible — especially with multiple editors on o... | true | ||
ueberdosis/tiptap | 7,924 | issue_to_patch | Explain conditional expression to clarify its purpose | ## Changes Overview
Refactor Server AI toolkit code to improve readability and clarify the meaning of the conditional expression.
<!-- Briefly describe your changes. -->
## Implementation Approach
Added an explanatory comment and simplified the conditional expression.
<!-- Describe your approach to imple... | 27b0eab3ff54cd06c74abd3ec0556e31f0c40705 | c758537baf5fb5507ca324fdccb208bc33e5c6ed | diff --git a/packages/server-ai-toolkit/src/hash-extension/server-ai-toolkit-hash-extension.ts b/packages/server-ai-toolkit/src/hash-extension/server-ai-toolkit-hash-extension.ts
index dbb351bb2d..9b8d26ea45 100644
--- a/packages/server-ai-toolkit/src/hash-extension/server-ai-toolkit-hash-extension.ts
+++ b/packages/se... | [
"packages/server-ai-toolkit/src/hash-extension/server-ai-toolkit-hash-extension.ts"
] | [] | true | ||
ueberdosis/tiptap | 7,912 | issue_to_patch | Publish a new stable version | This PR was opened by the [Changesets release](https://github.com/changesets/action) GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will ... | 7fb19eb8d4a9a013b7a40d07dddffe09eea3410b | 24773515c4f312f06ec827fa685a096a590bae30 | diff --git a/.changeset/2026-05-29-blockquote-backspace-lift-and-merge.md b/.changeset/2026-05-29-blockquote-backspace-lift-and-merge.md
deleted file mode 100644
index 59155175ab..0000000000
--- a/.changeset/2026-05-29-blockquote-backspace-lift-and-merge.md
+++ /dev/null
@@ -1,5 +0,0 @@
----
-'@tiptap/extension-blockqu... | [
".changeset/2026-05-29-blockquote-backspace-lift-and-merge.md",
".changeset/fix-server-ai-toolkit-hash-marks.md",
".changeset/little-shrimps-explain.md",
"CHANGELOG.md",
"demos/CHANGELOG.md",
"demos/package.json",
"packages-deprecated/extension-character-count/CHANGELOG.md",
"packages-deprecated/exten... | [] | true | ||
ueberdosis/tiptap | 7,920 | issue_to_patch | Include demos in release metadata scan | ## Changes Overview
Fixed issue that caused the `Publish` workflow to fail after commit https://github.com/ueberdosis/tiptap/commit/1c6619eccb6a9cdee46da3e8823a407acbe9e4d9
Fix the publish changelog aggregation script so it can resolve metadata for the private `demos` workspace package.
## Implementation Appro... | a7f9d0dd73feeca7e2c4e8a5a657484fa50198c8 | 43997f1627d1892d9efbbe55deb1114d37ff0f49 | diff --git a/scripts/aggregate-changeset.js b/scripts/aggregate-changeset.js
index 711ba95a4e..3861e4ffc9 100644
--- a/scripts/aggregate-changeset.js
+++ b/scripts/aggregate-changeset.js
@@ -112,10 +112,13 @@ async function addCommitsToChangesets(cwd, changesets) {
}
async function loadWorkspacePackages(cwd) {
- c... | [
"scripts/aggregate-changeset.js"
] | [] | true | ||
ueberdosis/tiptap | 7,918 | issue_to_patch | fix: only add hash attributes to nodes | Changes Overview
This PR fixes hash generation in `@tiptap/server-ai-toolkit` so marks are excluded from the hash-attribute registration logic. The package now only adds `_hash` to node extensions, which matches the intended behavior.
## Implementation Approach
I updated the global attribute filtering in `Serv... | ce85f6f04ba17d2ebbb1126165b94b94024483de | 6aab89b48e7c6bfa8019939a62fb858d658343af | diff --git a/.changeset/fix-server-ai-toolkit-hash-marks.md b/.changeset/fix-server-ai-toolkit-hash-marks.md
new file mode 100644
index 0000000000..9d38945975
--- /dev/null
+++ b/.changeset/fix-server-ai-toolkit-hash-marks.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/server-ai-toolkit': patch
+---
+
+Only add hash attributes to n... | [
".changeset/fix-server-ai-toolkit-hash-marks.md",
"packages/server-ai-toolkit/src/hash-extension/server-ai-toolkit-hash-extension.ts"
] | [] | true | ||
ueberdosis/tiptap | 7,907 | issue_to_patch | Selection extension shows incorrect highlight when selecting across multiple nodes
### Affected Packages
extensions/Selection
### Version(s)
3.x
### Bug Description
When using the [Selection extension](https://tiptap.dev/docs/editor/extensions/functionality/selection), the selection shows fine on blur when the se... | fix(extensions/selection): hide native selection while editor is blurred | ## Changes Overview
The `Selection` extension now hides the native browser selection while the editor is blurred, so only its styled `.selection` decoration is shown; previously a multi-line selection left the native highlight bleeding to the full width of each wrapped line, beyond the selected text.
## Implementatio... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | 75e840414c246dbfe3add8b645cb6becd7831fde | diff --git a/.changeset/little-shrimps-explain.md b/.changeset/little-shrimps-explain.md
new file mode 100644
index 0000000000..70035030eb
--- /dev/null
+++ b/.changeset/little-shrimps-explain.md
@@ -0,0 +1,6 @@
+---
+'@tiptap/extensions': patch
+'tiptap-demos': patch
+---
+
+Fix the `Selection` extension highlighting ... | [
".changeset/little-shrimps-explain.md",
"demos/src/Extensions/Selection/React/index.tsx",
"demos/src/Extensions/Selection/Vue/index.vue",
"packages/extensions/__tests__/selection.spec.ts",
"packages/extensions/src/selection/selection.ts"
] | [] | diff --git a/packages/extensions/__tests__/selection.spec.ts b/packages/extensions/__tests__/selection.spec.ts
new file mode 100644
index 0000000000..6456f40bee
--- /dev/null
+++ b/packages/extensions/__tests__/selection.spec.ts
@@ -0,0 +1,45 @@
+import { Editor } from '@tiptap/core'
+import Document from '@tiptap/exte... | true |
ueberdosis/tiptap | 7,891 | issue_to_patch | feat(extension-blockquote): lift and merge on backspace at child boundaries | ## Changes Overview
Make Backspace at the start of a blockquote child interactively split or merge the blockquote, matching the affordance Enter on a blank child already provides.
Before:
- Inside `<blockquote><p>A</p><p>B</p></blockquote>` with the caret before `B`, Backspace did nothing useful — `joinBackwar... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | 6b2f67a6ea65822862668d71eb800481ede1cd8c | diff --git a/.changeset/2026-05-29-blockquote-backspace-lift-and-merge.md b/.changeset/2026-05-29-blockquote-backspace-lift-and-merge.md
new file mode 100644
index 0000000000..59155175ab
--- /dev/null
+++ b/.changeset/2026-05-29-blockquote-backspace-lift-and-merge.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/extension-blockquote'... | [
".changeset/2026-05-29-blockquote-backspace-lift-and-merge.md",
"demos/src/Nodes/Blockquote/index.spec.ts",
"packages/extension-blockquote/src/blockquote.tsx",
"packages/extension-blockquote/src/handleBackspace.ts"
] | [
{
"comment": "The `targetPos = blockStart - 2` offset is a “magic number”, and the comment referencing “closing tags” is misleading for ProseMirror positions. Consider deriving `targetPos` via named intermediate positions (e.g., compute the end of the previous blockquote, then the end of its last textblock) and... | diff --git a/demos/src/Nodes/Blockquote/index.spec.ts b/demos/src/Nodes/Blockquote/index.spec.ts
index 468f7f0f73..be2bad8fcb 100644
--- a/demos/src/Nodes/Blockquote/index.spec.ts
+++ b/demos/src/Nodes/Blockquote/index.spec.ts
@@ -80,6 +80,70 @@ test.describe(`${demoPath}/${demoName}`, () => {
await page.keybo... | true | |
ueberdosis/tiptap | 7,891 | comment_to_fix | feat(extension-blockquote): lift and merge on backspace at child boundaries | The `targetPos = blockStart - 2` offset is a “magic number”, and the comment referencing “closing tags” is misleading for ProseMirror positions. Consider deriving `targetPos` via named intermediate positions (e.g., compute the end of the previous blockquote, then the end of its last textblock) and updating the comment ... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | 6b2f67a6ea65822862668d71eb800481ede1cd8c | diff --git a/packages/extension-blockquote/src/blockquote.tsx b/packages/extension-blockquote/src/blockquote.tsx
index 55ec225438..272cc9daa1 100644
--- a/packages/extension-blockquote/src/blockquote.tsx
+++ b/packages/extension-blockquote/src/blockquote.tsx
@@ -1,6 +1,8 @@
/** @jsxImportSource @tiptap/core */
import... | [
"packages/extension-blockquote/src/blockquote.tsx"
] | [
{
"comment": "The `targetPos = blockStart - 2` offset is a “magic number”, and the comment referencing “closing tags” is misleading for ProseMirror positions. Consider deriving `targetPos` via named intermediate positions (e.g., compute the end of the previous blockquote, then the end of its last textblock) and... | true | ||
ueberdosis/tiptap | 7,891 | comment_to_fix | feat(extension-blockquote): lift and merge on backspace at child boundaries | These tests use hard-coded document offsets (`from: 5`, `from: 6`) to place the cursor, which can become brittle if the schema, node structure, or serialization changes. Prefer computing the selection position from the current `doc` (e.g., resolve the position of the target text/node programmatically) so the tests asse... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | 6b2f67a6ea65822862668d71eb800481ede1cd8c | diff --git a/demos/src/Nodes/Blockquote/index.spec.ts b/demos/src/Nodes/Blockquote/index.spec.ts
index 468f7f0f73..be2bad8fcb 100644
--- a/demos/src/Nodes/Blockquote/index.spec.ts
+++ b/demos/src/Nodes/Blockquote/index.spec.ts
@@ -80,6 +80,70 @@ test.describe(`${demoPath}/${demoName}`, () => {
await page.keybo... | [
"demos/src/Nodes/Blockquote/index.spec.ts"
] | [
{
"comment": "These tests use hard-coded document offsets (`from: 5`, `from: 6`) to place the cursor, which can become brittle if the schema, node structure, or serialization changes. Prefer computing the selection position from the current `doc` (e.g., resolve the position of the target text/node programmatica... | true | ||
ueberdosis/tiptap | 7,891 | comment_to_fix | feat(extension-blockquote): lift and merge on backspace at child boundaries | I think extracting this into it's own function with a declarative name would be a good improvement. | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | 6b2f67a6ea65822862668d71eb800481ede1cd8c | diff --git a/packages/extension-blockquote/src/blockquote.tsx b/packages/extension-blockquote/src/blockquote.tsx
index 55ec225438..272cc9daa1 100644
--- a/packages/extension-blockquote/src/blockquote.tsx
+++ b/packages/extension-blockquote/src/blockquote.tsx
@@ -1,6 +1,8 @@
/** @jsxImportSource @tiptap/core */
import... | [
"packages/extension-blockquote/src/blockquote.tsx"
] | [
{
"comment": "I think extracting this into it's own function with a declarative name would be a good improvement.",
"path": "packages/extension-blockquote/src/blockquote.tsx",
"hunk": "@@ -126,6 +127,45 @@ export const Blockquote = Node.create<BlockquoteOptions>({\n addKeyboardShortcuts() {\n retu... | true | ||
ueberdosis/tiptap | 7,886 | issue_to_patch | Publish a new stable version | This PR was opened by the [Changesets release](https://github.com/changesets/action) GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will ... | 711c2e355123870220e5b51a9c580166bb9f6b4f | 5d5033665f36afe586256cb0fe09f1ba647f867d | diff --git a/.changeset/2026-05-29-file-handler-inline-image-drag-duplicate.md b/.changeset/2026-05-29-file-handler-inline-image-drag-duplicate.md
deleted file mode 100644
index e370b4258c..0000000000
--- a/.changeset/2026-05-29-file-handler-inline-image-drag-duplicate.md
+++ /dev/null
@@ -1,9 +0,0 @@
----
-'@tiptap/co... | [
".changeset/2026-05-29-file-handler-inline-image-drag-duplicate.md",
".changeset/2026-05-29-fix-nested-list-delete.md",
".changeset/2026-05-29-list-keymap-backspace-lift-first.md",
".changeset/2026-05-29-mark-clearable-unset-all-marks.md",
".changeset/2026-05-29-placeholder-empty-node-class-function.md",
... | [] | true | ||
ueberdosis/tiptap | 7,908 | issue_to_patch | fix(extension-unique-id): detach provider synced listener on destroy | The `synced` listener registered in `onCreate` was only removed once the event fired. If the editor was destroyed before the collaboration provider synced, the listener stayed attached to the shared provider, keeping the `createIds` closure — and the whole editor it captures — reachable and preventing garbage collectio... | b48e23f2f09715a106b81f1b6aa1439a2a559eb0 | af365bc5b88a70a036ab906d24188cc94d9d4b50 | diff --git a/.changeset/2026-06-02-fix-unique-id-synced-leak.md b/.changeset/2026-06-02-fix-unique-id-synced-leak.md
new file mode 100644
index 0000000000..fae76465ae
--- /dev/null
+++ b/.changeset/2026-06-02-fix-unique-id-synced-leak.md
@@ -0,0 +1,5 @@
+---
+"@tiptap/extension-unique-id": patch
+---
+
+Fix memory leak... | [
".changeset/2026-06-02-fix-unique-id-synced-leak.md",
"packages/extension-unique-id/__tests__/unique-id-collab.spec.ts",
"packages/extension-unique-id/src/unique-id.ts"
] | [] | diff --git a/packages/extension-unique-id/__tests__/unique-id-collab.spec.ts b/packages/extension-unique-id/__tests__/unique-id-collab.spec.ts
index 745fe6be21..bc14557914 100644
--- a/packages/extension-unique-id/__tests__/unique-id-collab.spec.ts
+++ b/packages/extension-unique-id/__tests__/unique-id-collab.spec.ts
@... | true | |
ueberdosis/tiptap | 7,906 | issue_to_patch | Bump prosemirror-tables so that findTable is always available on @tiptap/pm | ## Changes Overview
Bump `prosemirror-tables` in `@tiptap/pm` and the demos workspace so `findTable` is available, and align the related ProseMirror dependencies required by the newer tables package.
## Implementation Approach
Updated the dependency pins owned by `@tiptap/pm`, mirrored them in the demos workspace, r... | ae17f178e3e03b97b9292d1f8d82f73ae8623679 | 4025cfa1b0f0a51a3079e6cb48fa9fde3e2bdde5 | diff --git a/.changeset/2026-06-01-prosemirror-tables-findtable.md b/.changeset/2026-06-01-prosemirror-tables-findtable.md
new file mode 100644
index 0000000000..cbaca56cca
--- /dev/null
+++ b/.changeset/2026-06-01-prosemirror-tables-findtable.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/pm': patch
+---
+
+Bump `prosemirror-table... | [
".changeset/2026-06-01-prosemirror-tables-findtable.md",
"demos/package.json",
"demos/src/Examples/CustomDocument/index.spec.ts",
"demos/src/Extensions/TextAlign/index.spec.ts",
"packages/core/__tests__/insertContent.spec.ts",
"packages/extension-table/__tests__/tableCommands.spec.ts",
"packages/extensi... | [] | diff --git a/demos/src/Examples/CustomDocument/index.spec.ts b/demos/src/Examples/CustomDocument/index.spec.ts
index 4fd33e212b..03ef6f8384 100644
--- a/demos/src/Examples/CustomDocument/index.spec.ts
+++ b/demos/src/Examples/CustomDocument/index.spec.ts
@@ -67,7 +67,7 @@ test.describe(`${demoPath}/${demoName}`, () => ... | true | |
ueberdosis/tiptap | 7,740 | issue_to_patch | fix(collaboration-caret): guard against null awareness state values | ## Changes Overview
Guard against null/undefined awareness state values in `awarenessStatesToArray` to prevent runtime errors when a collaboration client disconnects.
## Implementation Approach
Added a null check on the `value` before spreading `value.user`. When `value` is falsy, only `clientId` is returned.
... | a4b939127821aac3a2139a467c5b257dd897da4e | 641c59d6b91fb1e85dc55bfe811fdffd42f1fe54 | diff --git a/.changeset/fix-collab-caret-null-awareness.md b/.changeset/fix-collab-caret-null-awareness.md
new file mode 100644
index 0000000000..d547b9b159
--- /dev/null
+++ b/.changeset/fix-collab-caret-null-awareness.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/extension-collaboration-caret': patch
+---
+
+Fix crash when aware... | [
".changeset/fix-collab-caret-null-awareness.md",
"packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts",
"packages/extension-collaboration-caret/src/collaboration-caret.ts"
] | [
{
"comment": "The `states` map is typed as `Map<number, Record<string, any> | null>` but the test also stores `undefined` entries, requiring `as any` casts. Consider widening the map value type to `Record<string, any> | null | undefined` (and removing the casts) so the test reflects the intended input shape mor... | diff --git a/packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts b/packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts
index fe6109281e..54f6d296d5 100644
--- a/packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts
+++ b/packages/extension-collabo... | true | |
ueberdosis/tiptap | 7,740 | comment_to_fix | fix(collaboration-caret): guard against null awareness state values | The `states` map is typed as `Map<number, Record<string, any> | null>` but the test also stores `undefined` entries, requiring `as any` casts. Consider widening the map value type to `Record<string, any> | null | undefined` (and removing the casts) so the test reflects the intended input shape more accurately.
```sugge... | a4b939127821aac3a2139a467c5b257dd897da4e | 641c59d6b91fb1e85dc55bfe811fdffd42f1fe54 | diff --git a/packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts b/packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts
index fe6109281e..54f6d296d5 100644
--- a/packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts
+++ b/packages/extension-collabo... | [
"packages/extension-collaboration-caret/__tests__/collaboration-caret.spec.ts"
] | [
{
"comment": "The `states` map is typed as `Map<number, Record<string, any> | null>` but the test also stores `undefined` entries, requiring `as any` casts. Consider widening the map value type to `Record<string, any> | null | undefined` (and removing the casts) so the test reflects the intended input shape mor... | true | ||
ueberdosis/tiptap | 7,740 | comment_to_fix | fix(collaboration-caret): guard against null awareness state values | Nit: `awarenessStatesToArray` still types `states` as `Map<number, Record<string, any>>`, but the new runtime logic explicitly supports `null`/`undefined` entries. Widening the parameter type to include `null | undefined` would make the guard meaningful in TypeScript and avoid future unsafe `value.user` access during r... | a4b939127821aac3a2139a467c5b257dd897da4e | 641c59d6b91fb1e85dc55bfe811fdffd42f1fe54 | diff --git a/packages/extension-collaboration-caret/src/collaboration-caret.ts b/packages/extension-collaboration-caret/src/collaboration-caret.ts
index 210bab6eac..dc40f66e08 100644
--- a/packages/extension-collaboration-caret/src/collaboration-caret.ts
+++ b/packages/extension-collaboration-caret/src/collaboration-ca... | [
"packages/extension-collaboration-caret/src/collaboration-caret.ts"
] | [
{
"comment": "Nit: `awarenessStatesToArray` still types `states` as `Map<number, Record<string, any>>`, but the new runtime logic explicitly supports `null`/`undefined` entries. Widening the parameter type to include `null | undefined` would make the guard meaningful in TypeScript and avoid future unsafe `value... | true | ||
ueberdosis/tiptap | 7,740 | comment_to_fix | fix(collaboration-caret): guard against null awareness state values | ```suggestion
if (value && value.user) {
``` | a4b939127821aac3a2139a467c5b257dd897da4e | 641c59d6b91fb1e85dc55bfe811fdffd42f1fe54 | diff --git a/packages/extension-collaboration-caret/src/collaboration-caret.ts b/packages/extension-collaboration-caret/src/collaboration-caret.ts
index 210bab6eac..dc40f66e08 100644
--- a/packages/extension-collaboration-caret/src/collaboration-caret.ts
+++ b/packages/extension-collaboration-caret/src/collaboration-ca... | [
"packages/extension-collaboration-caret/src/collaboration-caret.ts"
] | [
{
"comment": "```suggestion\n if (value && value.user) {\n```",
"path": "packages/extension-collaboration-caret/src/collaboration-caret.ts",
"hunk": "@@ -82,11 +82,16 @@ declare module '@tiptap/core' {\n }\n }\n \n-const awarenessStatesToArray = (states: Map<number, Record<string, any>>) => {\n+con... | true | ||
ueberdosis/tiptap | 7,903 | issue_to_patch | renderJSONContentToReactElement Type mismatch
### Affected Packages
@tiptap/static-renderer
### Version(s)
3.10.2
### Bug Description
In node mapping, there is no `node.text` inside the text node. Also, every prop of the node, e.g., type, is `any`.
`Property 'text' does not exist on type 'NodeType'.ts(2339)`
Th... | fix(static-renderer): type JSON renderers so JSONContent works without casts | ## Changes Overview
`renderJSONContentToReactElement` and `renderJSONContentToString` defaulted their node/mark generics to the ProseMirror-oriented `NodeType`/`MarkType`. As a result, in JSON node mappings `node.text` errored, every node prop was typed `any`, and passing a value typed as `JSONContent` failed (`type` ... | 45237e718ac9e42103de0490126c141cfa5a5741 | d0434a6542abb02b199577f0c430decf3ea826c7 | diff --git a/.changeset/chatty-rules-occur.md b/.changeset/chatty-rules-occur.md
new file mode 100644
index 0000000000..b4ff1bd5d1
--- /dev/null
+++ b/.changeset/chatty-rules-occur.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/static-renderer': patch
+---
+
+Fix the types of the JSON static renderers (`renderJSONContentToReactElem... | [
".changeset/chatty-rules-occur.md",
"packages/static-renderer/__tests__/json-string.spec.ts",
"packages/static-renderer/src/json/html-string/string.ts",
"packages/static-renderer/src/json/react/react.ts",
"packages/static-renderer/src/json/renderer.ts"
] | [
{
"comment": "It's a quite long changeset - I would reduce it to something that describes quickly to the user what changed for them. I don't think they care about what exact types were changed around but rather what improved/changed for them directly.",
"path": ".changeset/chatty-rules-occur.md",
"hunk"... | diff --git a/packages/static-renderer/__tests__/json-string.spec.ts b/packages/static-renderer/__tests__/json-string.spec.ts
index abae572c5d..276bf74a40 100644
--- a/packages/static-renderer/__tests__/json-string.spec.ts
+++ b/packages/static-renderer/__tests__/json-string.spec.ts
@@ -1,4 +1,3 @@
-import type { TextTy... | true |
ueberdosis/tiptap | 7,903 | comment_to_fix | fix(static-renderer): type JSON renderers so JSONContent works without casts | It's a quite long changeset - I would reduce it to something that describes quickly to the user what changed for them. I don't think they care about what exact types were changed around but rather what improved/changed for them directly. | 45237e718ac9e42103de0490126c141cfa5a5741 | d0434a6542abb02b199577f0c430decf3ea826c7 | diff --git a/.changeset/chatty-rules-occur.md b/.changeset/chatty-rules-occur.md
new file mode 100644
index 0000000000..b4ff1bd5d1
--- /dev/null
+++ b/.changeset/chatty-rules-occur.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/static-renderer': patch
+---
+
+Fix the types of the JSON static renderers (`renderJSONContentToReactElem... | [
".changeset/chatty-rules-occur.md"
] | [
{
"comment": "It's a quite long changeset - I would reduce it to something that describes quickly to the user what changed for them. I don't think they care about what exact types were changed around but rather what improved/changed for them directly.",
"path": ".changeset/chatty-rules-occur.md",
"hunk"... | true | ||
ueberdosis/tiptap | 7,888 | issue_to_patch | [BULLET LIST] Instead of moving up the items of the list, the items get deleted on pressing DEL.
### Affected Packages
extension-bullet-list
### Version(s)
2.0.0, 3.0.0
### Description of the Bug
Instead of moving up the items of the list, the items get deleted on pressing DEL.
The error occurs when the user pres... | fix(extensions/list): handle deletion of nested list items correctly | ## Changes Overview
Fix handling of nested list items when pressing the Delete key at the beginning of a list item with a branching nested sublist. Nested items are hoisted to the parent list instead of being node-selected and deleted on the next keypress.
## Implementation Approach
Added three new helper functions ... | ec291dd2a67709c5f34164fe4f1b228e1a4cf3d1 | 7d94b7500e3f802d28e72060dc65272057eb58bd | diff --git a/.changeset/2026-05-29-fix-nested-list-delete.md b/.changeset/2026-05-29-fix-nested-list-delete.md
new file mode 100644
index 0000000000..fefcfcc6b2
--- /dev/null
+++ b/.changeset/2026-05-29-fix-nested-list-delete.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/extension-list': patch
+---
+
+Fix delete at the end of a li... | [
".changeset/2026-05-29-fix-nested-list-delete.md",
"packages/extension-list/__tests__/listItemDelete.spec.ts",
"packages/extension-list/src/helpers/createBranchingListDeleteKeymap.ts",
"packages/extension-list/src/helpers/getBranchingNestedListAtCursor.ts",
"packages/extension-list/src/helpers/handleDeleteB... | [
{
"comment": "These re-exports break the existing alphabetical ordering of this file (note `handleBackspace`, `handleDelete`, `hasListBefore`, ... are sorted by filename). The newly added entries (`handleDeleteBranchingNestedList`, `hasBranchingNestedListAfterCursor`, `hoistBranchingNestedList`) are not in alph... | diff --git a/packages/extension-list/__tests__/listItemDelete.spec.ts b/packages/extension-list/__tests__/listItemDelete.spec.ts
new file mode 100644
index 0000000000..cb0b0a769c
--- /dev/null
+++ b/packages/extension-list/__tests__/listItemDelete.spec.ts
@@ -0,0 +1,230 @@
+import { Editor } from '@tiptap/core'
+import... | true |
ueberdosis/tiptap | 7,888 | comment_to_fix | fix(extensions/list): handle deletion of nested list items correctly | `handleDeleteBranchingNestedList` is defined and exported here (and re-exported from both `helpers/index.ts` and `keymap/listHelpers/index.ts`), but nothing in the codebase calls it — the new keymap handlers in `list-item.ts` and `task-item.ts` invoke `hoistBranchingNestedList` directly via `editor.commands.command(...... | ec291dd2a67709c5f34164fe4f1b228e1a4cf3d1 | 7d94b7500e3f802d28e72060dc65272057eb58bd | diff --git a/packages/extension-list/src/helpers/handleDeleteBranchingNestedList.ts b/packages/extension-list/src/helpers/handleDeleteBranchingNestedList.ts
new file mode 100644
index 0000000000..70d39c3ffb
--- /dev/null
+++ b/packages/extension-list/src/helpers/handleDeleteBranchingNestedList.ts
@@ -0,0 +1,25 @@
+impo... | [
"packages/extension-list/src/helpers/handleDeleteBranchingNestedList.ts"
] | [
{
"comment": "`handleDeleteBranchingNestedList` is defined and exported here (and re-exported from both `helpers/index.ts` and `keymap/listHelpers/index.ts`), but nothing in the codebase calls it — the new keymap handlers in `list-item.ts` and `task-item.ts` invoke `hoistBranchingNestedList` directly via `edito... | true | ||
ueberdosis/tiptap | 7,888 | comment_to_fix | fix(extensions/list): handle deletion of nested list items correctly | This is a behavior change in default list keyboard handling (Delete / Mod-Delete on `listItem` and nested `taskItem`) introduced to fix issue #6906, but no unit tests are added. The package already has a Vitest setup (`packages/extension-list/__tests__/`) with editor fixtures suitable for asserting transaction outputs ... | ec291dd2a67709c5f34164fe4f1b228e1a4cf3d1 | 7d94b7500e3f802d28e72060dc65272057eb58bd | diff --git a/packages/extension-list/src/helpers/hoistBranchingNestedList.ts b/packages/extension-list/src/helpers/hoistBranchingNestedList.ts
new file mode 100644
index 0000000000..792f6bd921
--- /dev/null
+++ b/packages/extension-list/src/helpers/hoistBranchingNestedList.ts
@@ -0,0 +1,56 @@
+import { Fragment } from ... | [
"packages/extension-list/src/helpers/hoistBranchingNestedList.ts"
] | [
{
"comment": "This is a behavior change in default list keyboard handling (Delete / Mod-Delete on `listItem` and nested `taskItem`) introduced to fix issue #6906, but no unit tests are added. The package already has a Vitest setup (`packages/extension-list/__tests__/`) with editor fixtures suitable for assertin... | true | ||
ueberdosis/tiptap | 7,888 | comment_to_fix | fix(extensions/list): handle deletion of nested list items correctly | `hoistBranchingNestedList` calls `hasBranchingNestedListAfterCursor(...)` and then independently re-derives the list-item depth (via the local `findListItemDepth`), the nested list position (`$from.after()`), and re-checks `wrapperNames.includes(nodeAfter.type.name)`. The work `hasBranchingNestedListAfterCursor` alread... | ec291dd2a67709c5f34164fe4f1b228e1a4cf3d1 | 7d94b7500e3f802d28e72060dc65272057eb58bd | diff --git a/packages/extension-list/src/helpers/hoistBranchingNestedList.ts b/packages/extension-list/src/helpers/hoistBranchingNestedList.ts
new file mode 100644
index 0000000000..792f6bd921
--- /dev/null
+++ b/packages/extension-list/src/helpers/hoistBranchingNestedList.ts
@@ -0,0 +1,56 @@
+import { Fragment } from ... | [
"packages/extension-list/src/helpers/hoistBranchingNestedList.ts"
] | [
{
"comment": "`hoistBranchingNestedList` calls `hasBranchingNestedListAfterCursor(...)` and then independently re-derives the list-item depth (via the local `findListItemDepth`), the nested list position (`$from.after()`), and re-checks `wrapperNames.includes(nodeAfter.type.name)`. The work `hasBranchingNestedL... | true | ||
ueberdosis/tiptap | 7,898 | issue_to_patch | Markdown extension misinterprets some escaped characters as formatting
### Affected Packages
@tiptap/markdown
### Version(s)
3.10.7
### Bug Description
## Issue
When using a Markdown editor, if a user inputs two escape characters, the editor appears to interpret this as a request to format any text in between th... | fix(markdown): add backslash-escape handling in parser and serializer | ## Changes Overview
Fix backslash-escape handling in `@tiptap/markdown`. Backslash-escaped markdown characters (e.g. `\*`, `\_`, `\\`) were silently dropped during parsing, and markdown-significant characters in text nodes were not escaped during serialization.
## Implementation Approach
Two-pronged fix in `Markdown... | 3d4f94c4bfd68580a49af155523b898c1eb3efd5 | 4b9bfd36f75d85e6ee0ad99901fc3293a9fd9ec8 | diff --git a/.changeset/2026-06-01-attrs-marks-equal-utilities.md b/.changeset/2026-06-01-attrs-marks-equal-utilities.md
new file mode 100644
index 0000000000..a3edb62221
--- /dev/null
+++ b/.changeset/2026-06-01-attrs-marks-equal-utilities.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/core': patch
+---
+
+Add `attrsEqual` and `ma... | [
".changeset/2026-06-01-attrs-marks-equal-utilities.md",
".changeset/fix-markdown-escape-handling.md",
"packages/core/__tests__/attrsEqual.spec.ts",
"packages/core/__tests__/marksEqual.spec.ts",
"packages/core/src/utilities/attrsEqual.ts",
"packages/core/src/utilities/index.ts",
"packages/core/src/utilit... | [
{
"comment": "Isn't better to do object equality check instead of string checks? This can fail for logically equal objects with different key order 🤔 also it allocates much more memory due to string serialization. Is there a reason we can't go with something like `lodash/isEqual`, or `fast-deep-equal`, or sim... | diff --git a/packages/core/__tests__/attrsEqual.spec.ts b/packages/core/__tests__/attrsEqual.spec.ts
new file mode 100644
index 0000000000..d5eb992be3
--- /dev/null
+++ b/packages/core/__tests__/attrsEqual.spec.ts
@@ -0,0 +1,53 @@
+import { attrsEqual } from '@tiptap/core'
+import { describe, expect, it } from 'vitest'... | true |
ueberdosis/tiptap | 7,898 | comment_to_fix | fix(markdown): add backslash-escape handling in parser and serializer | Isn't better to do object equality check instead of string checks? This can fail for logically equal objects with different key order 🤔 also it allocates much more memory due to string serialization. Is there a reason we can't go with something like `lodash/isEqual`, or `fast-deep-equal`, or similar instead? | 3d4f94c4bfd68580a49af155523b898c1eb3efd5 | 4b9bfd36f75d85e6ee0ad99901fc3293a9fd9ec8 | diff --git a/packages/markdown/src/MarkdownManager.ts b/packages/markdown/src/MarkdownManager.ts
index 3777729318..0d1c9fd5c8 100644
--- a/packages/markdown/src/MarkdownManager.ts
+++ b/packages/markdown/src/MarkdownManager.ts
@@ -10,6 +10,7 @@ import {
type MarkdownToken,
type MarkdownTokenizer,
type RenderCo... | [
"packages/markdown/src/MarkdownManager.ts"
] | [
{
"comment": "Isn't better to do object equality check instead of string checks? This can fail for logically equal objects with different key order 🤔 also it allocates much more memory due to string serialization. Is there a reason we can't go with something like `lodash/isEqual`, or `fast-deep-equal`, or sim... | true | ||
ueberdosis/tiptap | 7,901 | issue_to_patch | [Bug]: `VueNodeViewRenderer` with a NodeView that doesn't have a static `NodeViewContent` will result in overriding the entire DOM with the children
### Which packages did you experience the bug in?
vue-3
### What Tiptap version are you using?
2.0.2
### What’s the bug you are facing?
This line:
https://github.co... | fix(vue): prevent DOM thrashing when contentDOM is null | ## Changes Overview
Fix `VueNodeViewRenderer` thrashing the DOM when a non-leaf NodeView component does not include `<NodeViewContent>` or renders it conditionally. This matches the behavior of `ReactNodeViewRenderer`.
## Implementation Approach
The root cause was twofold:
1. **`contentDOM` returning `null` for non... | 3163b136e295d2144e9afe24f3c2c6f7dc1eb9a4 | 1283a1e89a48941f155e4182f98c6cb9747db4a5 | diff --git a/.changeset/fix-vuenodeview-contentdom-null-thrashing.md b/.changeset/fix-vuenodeview-contentdom-null-thrashing.md
new file mode 100644
index 0000000000..a44de2f1fa
--- /dev/null
+++ b/.changeset/fix-vuenodeview-contentdom-null-thrashing.md
@@ -0,0 +1,6 @@
+---
+'@tiptap/vue-3': patch
+'@tiptap/vue-2': patc... | [
".changeset/fix-vuenodeview-contentdom-null-thrashing.md",
"packages/vue-2/src/NodeViewContent.ts",
"packages/vue-2/src/VueNodeViewRenderer.ts",
"packages/vue-3/__tests__/VueNodeViewRenderer.spec.ts",
"packages/vue-3/src/NodeViewContent.ts",
"packages/vue-3/src/VueNodeViewRenderer.ts"
] | [
{
"comment": "In Vue 2 `VueRenderer` invokes `$mount()` synchronously in its constructor (`packages/vue-2/src/VueRenderer.ts:14`), so by the time `new VueRenderer(Component, ...)` returns at line 134, the component tree (including any `NodeViewContent`) has already mounted and `nodeViewContentRef` has been invo... | diff --git a/packages/vue-3/__tests__/VueNodeViewRenderer.spec.ts b/packages/vue-3/__tests__/VueNodeViewRenderer.spec.ts
new file mode 100644
index 0000000000..afb61d90e5
--- /dev/null
+++ b/packages/vue-3/__tests__/VueNodeViewRenderer.spec.ts
@@ -0,0 +1,137 @@
+import { Editor, Node, mergeAttributes } from '@tiptap/co... | true |
ueberdosis/tiptap | 7,901 | comment_to_fix | fix(vue): prevent DOM thrashing when contentDOM is null | In Vue 2 `VueRenderer` invokes `$mount()` synchronously in its constructor (`packages/vue-2/src/VueRenderer.ts:14`), so by the time `new VueRenderer(Component, ...)` returns at line 134, the component tree (including any `NodeViewContent`) has already mounted and `nodeViewContentRef` has been invoked — at that point `t... | 3163b136e295d2144e9afe24f3c2c6f7dc1eb9a4 | 1283a1e89a48941f155e4182f98c6cb9747db4a5 | diff --git a/packages/vue-2/src/VueNodeViewRenderer.ts b/packages/vue-2/src/VueNodeViewRenderer.ts
index 5f16ebd450..776ff8f299 100644
--- a/packages/vue-2/src/VueNodeViewRenderer.ts
+++ b/packages/vue-2/src/VueNodeViewRenderer.ts
@@ -47,6 +47,15 @@ class VueNodeView extends NodeView<Vue | VueConstructor, Editor, VueNo... | [
"packages/vue-2/src/VueNodeViewRenderer.ts"
] | [
{
"comment": "In Vue 2 `VueRenderer` invokes `$mount()` synchronously in its constructor (`packages/vue-2/src/VueRenderer.ts:14`), so by the time `new VueRenderer(Component, ...)` returns at line 134, the component tree (including any `NodeViewContent`) has already mounted and `nodeViewContentRef` has been invo... | true | ||
ueberdosis/tiptap | 7,901 | comment_to_fix | fix(vue): prevent DOM thrashing when contentDOM is null | The inner `contentDOMElement` is tagged with the same `data-node-view-content` attribute as the `NodeViewContent` component’s own element, producing two nested elements with that attribute when the user renders `<node-view-content>`. User-authored CSS selectors targeting `[data-node-view-content]` will match both, and ... | 3163b136e295d2144e9afe24f3c2c6f7dc1eb9a4 | 1283a1e89a48941f155e4182f98c6cb9747db4a5 | diff --git a/packages/vue-3/src/VueNodeViewRenderer.ts b/packages/vue-3/src/VueNodeViewRenderer.ts
index e87df2d041..3b73e4d036 100644
--- a/packages/vue-3/src/VueNodeViewRenderer.ts
+++ b/packages/vue-3/src/VueNodeViewRenderer.ts
@@ -81,6 +81,19 @@ class VueNodeView extends NodeView<Component, Editor, VueNodeViewRende... | [
"packages/vue-3/src/VueNodeViewRenderer.ts"
] | [
{
"comment": "The inner `contentDOMElement` is tagged with the same `data-node-view-content` attribute as the `NodeViewContent` component’s own element, producing two nested elements with that attribute when the user renders `<node-view-content>`. User-authored CSS selectors targeting `[data-node-view-content]`... | true | ||
ueberdosis/tiptap | 7,901 | comment_to_fix | fix(vue): prevent DOM thrashing when contentDOM is null | In vue-3, `inject: ['nodeViewContentRef']` is declared without a default value. When `NodeViewContent` is rendered outside a Tiptap node-view context (e.g. tests, Storybook, accidental misuse), Vue 3 will emit an "injection not found" warning. The vue-2 sibling correctly uses `{ default: undefined }` (`packages/vue-2/s... | 3163b136e295d2144e9afe24f3c2c6f7dc1eb9a4 | 1283a1e89a48941f155e4182f98c6cb9747db4a5 | diff --git a/packages/vue-3/src/NodeViewContent.ts b/packages/vue-3/src/NodeViewContent.ts
index 65dc885b8e..6912b33e62 100644
--- a/packages/vue-3/src/NodeViewContent.ts
+++ b/packages/vue-3/src/NodeViewContent.ts
@@ -10,6 +10,24 @@ export const NodeViewContent = defineComponent({
},
},
+ inject: {
+ nod... | [
"packages/vue-3/src/NodeViewContent.ts"
] | [
{
"comment": "In vue-3, `inject: ['nodeViewContentRef']` is declared without a default value. When `NodeViewContent` is rendered outside a Tiptap node-view context (e.g. tests, Storybook, accidental misuse), Vue 3 will emit an \"injection not found\" warning. The vue-2 sibling correctly uses `{ default: undefin... | true | ||
ueberdosis/tiptap | 7,900 | issue_to_patch | [Bug]: Inserting inline code removes preceding non-whitespace characters
### Affected Packages
extensions-code
### Version(s)
2.12.0
### Bug Description
When inserting inline code with a preceding non-whitespace character, the preceding character gets removed.
To reproduce, type `a`, then backtick, then `code`, ... | fix(extension/code): preserve preceding characters when using inline code markdown | ## Changes Overview
Typing a character before an inline code markdown shortcut (e.g. `a`code``) would delete the preceding character. This is now fixed — the preceding character is preserved.
## Implementation Approach
The input and paste rule regexes used a capturing group `(^|[^`])` to check that the preceding cha... | a61046fdb0b2d3d5fcacae6427e9329d736ed325 | 05591325bfb7fa58bb47eb921cded58fec75269a | diff --git a/.changeset/fix-code-input-rule-preceding-char.md b/.changeset/fix-code-input-rule-preceding-char.md
new file mode 100644
index 0000000000..b7061936d9
--- /dev/null
+++ b/.changeset/fix-code-input-rule-preceding-char.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/extension-code': patch
+---
+
+Fixed a bug where typing a... | [
".changeset/fix-code-input-rule-preceding-char.md",
"demos/src/Marks/Code/index.spec.ts",
"packages/extension-code/__tests__/code.spec.ts",
"packages/extension-code/src/code.ts"
] | [
{
"comment": "`inputRegex` is exported from `@tiptap/extension-code` via `src/index.ts`, so changing it from a `RegExp` to a function is a public API/type break for consumers that import it or use RegExp APIs such as `test`/`toMatch`. Keep the existing `RegExp` export and use a separately named internal finder ... | diff --git a/demos/src/Marks/Code/index.spec.ts b/demos/src/Marks/Code/index.spec.ts
index 4f1b21be3b..cc108fcc04 100644
--- a/demos/src/Marks/Code/index.spec.ts
+++ b/demos/src/Marks/Code/index.spec.ts
@@ -57,7 +57,25 @@ test.describe(`${demoPath}/${demoName}`, () => {
const editor = await getEditor(page)
... | true |
ueberdosis/tiptap | 7,900 | comment_to_fix | fix(extension/code): preserve preceding characters when using inline code markdown | `inputRegex` is exported from `@tiptap/extension-code` via `src/index.ts`, so changing it from a `RegExp` to a function is a public API/type break for consumers that import it or use RegExp APIs such as `test`/`toMatch`. Keep the existing `RegExp` export and use a separately named internal finder for the input rule, or... | a61046fdb0b2d3d5fcacae6427e9329d736ed325 | 05591325bfb7fa58bb47eb921cded58fec75269a | diff --git a/packages/extension-code/src/code.ts b/packages/extension-code/src/code.ts
index 14a276c920..1f317bfbfc 100644
--- a/packages/extension-code/src/code.ts
+++ b/packages/extension-code/src/code.ts
@@ -1,3 +1,4 @@
+import type { InputRuleMatch, PasteRuleMatch } from '@tiptap/core'
import { Mark, markInputRule... | [
"packages/extension-code/src/code.ts"
] | [
{
"comment": "`inputRegex` is exported from `@tiptap/extension-code` via `src/index.ts`, so changing it from a `RegExp` to a function is a public API/type break for consumers that import it or use RegExp APIs such as `test`/`toMatch`. Keep the existing `RegExp` export and use a separately named internal finder ... | true | ||
ueberdosis/tiptap | 7,900 | comment_to_fix | fix(extension/code): preserve preceding characters when using inline code markdown | `pasteRegex` is also a public export, and this change replaces the previous `RegExp` value with a function returning match objects. That breaks existing consumers that pass the exported regex to string/RegExp APIs; please preserve the exported regex and introduce a separate paste-rule finder for the extension, or relea... | a61046fdb0b2d3d5fcacae6427e9329d736ed325 | 05591325bfb7fa58bb47eb921cded58fec75269a | diff --git a/packages/extension-code/src/code.ts b/packages/extension-code/src/code.ts
index 14a276c920..1f317bfbfc 100644
--- a/packages/extension-code/src/code.ts
+++ b/packages/extension-code/src/code.ts
@@ -1,3 +1,4 @@
+import type { InputRuleMatch, PasteRuleMatch } from '@tiptap/core'
import { Mark, markInputRule... | [
"packages/extension-code/src/code.ts"
] | [
{
"comment": "`pasteRegex` is also a public export, and this change replaces the previous `RegExp` value with a function returning match objects. That breaks existing consumers that pass the exported regex to string/RegExp APIs; please preserve the exported regex and introduce a separate paste-rule finder for t... | true | ||
ueberdosis/tiptap | 7,900 | comment_to_fix | fix(extension/code): preserve preceding characters when using inline code markdown | This Markdown example is parsed as an inline code span for only `a`, leaving the backticks around `code` outside the code span in the generated changelog. Escape or wrap the full shortcut with a longer code fence so the rendered changeset shows the intended `a` followed by backtick-delimited code. | a61046fdb0b2d3d5fcacae6427e9329d736ed325 | 05591325bfb7fa58bb47eb921cded58fec75269a | diff --git a/.changeset/fix-code-input-rule-preceding-char.md b/.changeset/fix-code-input-rule-preceding-char.md
new file mode 100644
index 0000000000..b7061936d9
--- /dev/null
+++ b/.changeset/fix-code-input-rule-preceding-char.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/extension-code': patch
+---
+
+Fixed a bug where typing a... | [
".changeset/fix-code-input-rule-preceding-char.md"
] | [
{
"comment": "This Markdown example is parsed as an inline code span for only `a`, leaving the backticks around `code` outside the code span in the generated changelog. Escape or wrap the full shortcut with a longer code fence so the rendered changeset shows the intended `a` followed by backtick-delimited code.... | true | ||
ueberdosis/tiptap | 7,887 | issue_to_patch | [PRO]: unsetAllMarks deletes comments
### Affected Packages
extension-comments
### Version(s)
2.17.5
### Description of the Bug
Our Tiptap editor has a "clear formatting" button that calls [`unsetAllMarks`](https://tiptap.dev/docs/editor/api/commands/nodes-and-marks/unset-all-marks) to remove formatting from the ... | fix(core): add clearable mark option for unsetAllMarks | ## Changes Overview
Adds a `clearable` mark extension option (default `true`) and updates `unsetAllMarks` to only remove clearable marks. This fixes bulk "clear formatting" unintentionally removing semantic marks such as comments (see #6348).
Pro packages can opt out by setting `clearable: false` on comment/sugge... | aff939d846e49ede2fd5fffe5ffa80ee9259428a | 5d0cea34cd0d0c2cd882889af7a912824fa72ebf | diff --git a/.changeset/2026-05-29-mark-clearable-unset-all-marks.md b/.changeset/2026-05-29-mark-clearable-unset-all-marks.md
new file mode 100644
index 0000000000..aabe90d19f
--- /dev/null
+++ b/.changeset/2026-05-29-mark-clearable-unset-all-marks.md
@@ -0,0 +1,5 @@
+---
+'@tiptap/core': patch
+---
+
+Add `clearable`... | [
".changeset/2026-05-29-mark-clearable-unset-all-marks.md",
"packages/core/__tests__/unsetAllMarks.spec.ts",
"packages/core/src/ExtensionManager.ts",
"packages/core/src/Mark.ts",
"packages/core/src/commands/unsetAllMarks.ts"
] | [
{
"comment": "This public option description mentions “similar bulk-clear commands,” but the new option is only consumed by `unsetAllMarks` in core. Narrowing the wording avoids implying other commands already respect `clearable`.",
"path": "packages/core/src/Mark.ts",
"hunk": "@@ -30,6 +30,13 @@ export... | diff --git a/packages/core/__tests__/unsetAllMarks.spec.ts b/packages/core/__tests__/unsetAllMarks.spec.ts
new file mode 100644
index 0000000000..a4d5afa31c
--- /dev/null
+++ b/packages/core/__tests__/unsetAllMarks.spec.ts
@@ -0,0 +1,107 @@
+import { Editor, Mark } from '@tiptap/core'
+import Bold from '@tiptap/extensi... | true |
ueberdosis/tiptap | 7,887 | comment_to_fix | fix(core): add clearable mark option for unsetAllMarks | This public option description mentions “similar bulk-clear commands,” but the new option is only consumed by `unsetAllMarks` in core. Narrowing the wording avoids implying other commands already respect `clearable`. | aff939d846e49ede2fd5fffe5ffa80ee9259428a | 5d0cea34cd0d0c2cd882889af7a912824fa72ebf | diff --git a/packages/core/src/Mark.ts b/packages/core/src/Mark.ts
index be2fc93141..cf4a78f2f6 100644
--- a/packages/core/src/Mark.ts
+++ b/packages/core/src/Mark.ts
@@ -30,6 +30,13 @@ export interface MarkConfig<Options = any, Storage = any> extends ExtendableConf
*/
keepOnSplit?: boolean | (() => boolean)
+... | [
"packages/core/src/Mark.ts"
] | [
{
"comment": "This public option description mentions “similar bulk-clear commands,” but the new option is only consumed by `unsetAllMarks` in core. Narrowing the wording avoids implying other commands already respect `clearable`.",
"path": "packages/core/src/Mark.ts",
"hunk": "@@ -30,6 +30,13 @@ export... | true | ||
ueberdosis/tiptap | 7,887 | comment_to_fix | fix(core): add clearable mark option for unsetAllMarks | I wonder if there could be an option to unset unclearable marks too, that defaults to `false` | aff939d846e49ede2fd5fffe5ffa80ee9259428a | 5d0cea34cd0d0c2cd882889af7a912824fa72ebf | diff --git a/packages/core/src/commands/unsetAllMarks.ts b/packages/core/src/commands/unsetAllMarks.ts
index 91a28962ed..4e3ce9a16c 100644
--- a/packages/core/src/commands/unsetAllMarks.ts
+++ b/packages/core/src/commands/unsetAllMarks.ts
@@ -4,17 +4,21 @@ declare module '@tiptap/core' {
interface Commands<ReturnTyp... | [
"packages/core/src/commands/unsetAllMarks.ts"
] | [
{
"comment": "I wonder if there could be an option to unset unclearable marks too, that defaults to `false`",
"path": "packages/core/src/commands/unsetAllMarks.ts",
"hunk": "@@ -1,30 +1,64 @@\n+import type { Node as ProseMirrorNode } from '@tiptap/pm/model'\n+import type { SelectionRange } from '@tiptap... | true |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.