From bfa9f40bdbf88c37c87b257e70f0e78a39f147a9 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 13:45:22 -0500
Subject: [PATCH 01/12] feat: Add data structure for tracking Jinja blocks

---
 dbt_meshify/storage/jinja_blocks.py | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 dbt_meshify/storage/jinja_blocks.py

diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
new file mode 100644
index 0000000..f88221e
--- /dev/null
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -0,0 +1,14 @@
+from dataclasses import dataclass
+from pathlib import Path
+
+
+@dataclass
+class JinjaBlock:
+    """
+    A common data structure for tracking blocks of text that represent Jinja blocks.
+    """
+
+    path: Path
+    start: int
+    end: int
+    content: str

From a2c8d79c029ee36d1f735c675df897e84bde18dc Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 16:30:02 -0500
Subject: [PATCH 02/12] feat: Add basic JinjaBlock parsing and associated unit
 tests

---
 dbt_meshify/dbt_projects.py                   | 15 ++++++
 dbt_meshify/storage/jinja_blocks.py           | 53 +++++++++++++++++++
 test-projects/split/split_proj/models/docs.md |  3 ++
 .../split_proj/models/marts/__models.yml      | 26 +++++----
 .../split_proj/models/staging/__models.yml    |  2 +-
 tests/unit/test_jinja_blocks.py               | 39 ++++++++++++++
 6 files changed, 127 insertions(+), 11 deletions(-)
 create mode 100644 test-projects/split/split_proj/models/docs.md
 create mode 100644 tests/unit/test_jinja_blocks.py

diff --git a/dbt_meshify/dbt_projects.py b/dbt_meshify/dbt_projects.py
index 5c34b88..bb22cb9 100644
--- a/dbt_meshify/dbt_projects.py
+++ b/dbt_meshify/dbt_projects.py
@@ -26,6 +26,7 @@
 
 from dbt_meshify.dbt import Dbt
 from dbt_meshify.exceptions import FatalMeshifyException
+from dbt_meshify.storage.jinja_blocks import JinjaBlock
 
 
 class BaseDbtProject:
@@ -311,6 +312,20 @@ def __init__(
 
         super().__init__(manifest, project, catalog, name, resources)
 
+        self.jinja_blocks: Dict[str, JinjaBlock] = self.find_jinja_blocks()
+
+    def find_jinja_blocks(self) -> Dict[str, JinjaBlock]:
+        """For a given dbt Project, find all Jinja blocks for docs and macros"""
+
+        blocks = {}
+
+        for unique_id, item in self.manifest.docs.items():
+            blocks[unique_id] = JinjaBlock.from_file(
+                path=self.path / item.original_file_path, block_type="docs", name=item.name
+            )
+
+        return blocks
+
     def select_resources(
         self,
         select: Optional[str] = None,
diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
index f88221e..dc78773 100644
--- a/dbt_meshify/storage/jinja_blocks.py
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -1,5 +1,7 @@
+import re
 from dataclasses import dataclass
 from pathlib import Path
+from typing import Tuple
 
 
 @dataclass
@@ -9,6 +11,57 @@ class JinjaBlock:
     """
 
     path: Path
+    block_type: str
+    name: str
     start: int
     end: int
     content: str
+
+    @staticmethod
+    def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int, int]:
+        """Find the line number that a block started."""
+        start_line = None
+        end_line = None
+
+        for match in re.finditer(
+            r"{%\s+" + block_type + r"\s+" + name + r"\s+%}", file_content, re.MULTILINE
+        ):
+            start = match.span()[0]  # .span() gives tuple (start, end)
+            start_line = file_content[:start].count("\n")
+            break
+
+        if start_line is None:
+            raise Exception(f"Unable to find a {block_type} block with the name {name}.")
+
+        for match in re.finditer(r"{%\s+end" + block_type + r"\s+%}", file_content, re.MULTILINE):
+            start = match.span()[0]  # .span() gives tuple (start, end)
+            new_end_line = file_content[:start].count("\n")
+
+            if new_end_line >= start_line:
+                end_line = new_end_line
+                break
+
+        if end_line is None:
+            raise Exception(f"Unable to find a the closing end{block_type} block for {name}.")
+
+        return start_line, end_line
+
+    @staticmethod
+    def isolate_content_from_line_range(file_content: str, start: int, end: int) -> str:
+        """Given content, a start line number, and an end line number, return the content of a Jinja block."""
+        print(file_content.split("\n")[start + 1 :])
+        return "/n".join(file_content.split("\n")[start + 1 : end])
+
+    @classmethod
+    def from_file(cls, path: Path, block_type: str, name: str) -> "JinjaBlock":
+        """Find a specific Jinja block within a file, based on the block type and the name."""
+
+        file_content = open(path).read()
+        start, end = cls.find_block_range(file_content, block_type, name)
+        content = cls.isolate_content_from_line_range(
+            file_content=file_content, start=start, end=end
+        )
+
+        return cls(
+            path=path, block_type=block_type, name=name, start=start, end=end, content=content
+        )
diff --git a/test-projects/split/split_proj/models/docs.md b/test-projects/split/split_proj/models/docs.md
new file mode 100644
index 0000000..491d656
--- /dev/null
+++ b/test-projects/split/split_proj/models/docs.md
@@ -0,0 +1,3 @@
+{% docs customer_id %}
+The unique key for each customer.
+{% enddocs %}
diff --git a/test-projects/split/split_proj/models/marts/__models.yml b/test-projects/split/split_proj/models/marts/__models.yml
index a9823fa..685b402 100644
--- a/test-projects/split/split_proj/models/marts/__models.yml
+++ b/test-projects/split/split_proj/models/marts/__models.yml
@@ -2,11 +2,12 @@ version: 2
 
 models:
   - name: customers
-    description: Customer overview data mart, offering key details for each unique
+    description:
+      Customer overview data mart, offering key details for each unique
       customer. One row per customer.
     columns:
       - name: customer_id
-        description: The unique key of the orders mart.
+        description: "{{ doc('customer_id') }}"
         tests:
           - not_null
           - unique
@@ -19,20 +20,24 @@ models:
       - name: last_ordered_at
         description: The timestamp of a customer's most recent order.
       - name: lifetime_spend_pretax
-        description: The sum of all the pre-tax subtotals of every order a customer
+        description:
+          The sum of all the pre-tax subtotals of every order a customer
           has placed.
       - name: lifetime_spend
-        description: The sum of all the order totals (including tax) that a customer
+        description:
+          The sum of all the order totals (including tax) that a customer
           has ever placed.
       - name: customer_type
-        description: Options are 'new' or 'returning', indicating if a customer has
+        description:
+          Options are 'new' or 'returning', indicating if a customer has
           ordered more than once or has only placed their first order to date.
         tests:
           - accepted_values:
               values: [new, returning]
 
   - name: orders
-    description: Order overview data mart, offering key details for each order inlcluding
+    description:
+      Order overview data mart, offering key details for each order inlcluding
       if it's a customer's first order and a food vs. drink item breakdown. One row
       per order.
     tests:
@@ -53,7 +58,8 @@ models:
               to: ref('stg_customers')
               field: customer_id
       - name: location_id
-        description: The foreign key relating to the location the order was placed
+        description:
+          The foreign key relating to the location the order was placed
           at.
       - name: order_total
         description: The total amount of the order in USD including tax.
@@ -74,14 +80,14 @@ models:
       - name: order_cost
         description: The sum of supply expenses to fulfill the order.
       - name: location_name
-        description: The full location name of where this order was placed. Denormalized
+        description:
+          The full location name of where this order was placed. Denormalized
           from `stg_locations`.
       - name: is_food_order
         description: A boolean indicating if this order included any food items.
       - name: is_drink_order
         description: A boolean indicating if this order included any drink items.
 
-
   - name: leaf_node
     description: A leaf node model that is not referenced by any other model.
     columns:
@@ -89,4 +95,4 @@ models:
         description: The unique key of the leaf node.
         tests:
           - not_null
-          - unique
\ No newline at end of file
+          - unique
diff --git a/test-projects/split/split_proj/models/staging/__models.yml b/test-projects/split/split_proj/models/staging/__models.yml
index 1667c52..125edb7 100644
--- a/test-projects/split/split_proj/models/staging/__models.yml
+++ b/test-projects/split/split_proj/models/staging/__models.yml
@@ -5,7 +5,7 @@ models:
     description: Customer data with basic cleaning and transformation applied, one row per customer.
     columns:
       - name: customer_id
-        description: The unique key for each customer.
+        description: "{{ doc('customer_id') }}"
         tests:
           - not_null
           - unique
diff --git a/tests/unit/test_jinja_blocks.py b/tests/unit/test_jinja_blocks.py
new file mode 100644
index 0000000..d638458
--- /dev/null
+++ b/tests/unit/test_jinja_blocks.py
@@ -0,0 +1,39 @@
+from dbt_meshify.storage.jinja_blocks import JinjaBlock
+
+string = """\
+
+
+{% docs customer_id %}
+The unique key for each customer.
+{% enddocs %}
+"""
+
+multiple_blocks = """\
+
+
+{% docs customer_id %}
+The unique key for each customer.
+{% enddocs %}
+
+{% docs potato_name %}
+The name of the customer's favorite potato dish.
+{% enddocs %}
+"""
+
+
+class TestJinjaBlock:
+    def test_from_file_detects_block_range(self):
+        range = JinjaBlock.find_block_range(string, "docs", "customer_id")
+        assert range == (2, 4)
+
+    def test_from_file_extracts_content(self):
+        content = JinjaBlock.isolate_content_from_line_range(string, 2, 4)
+        assert content == "The unique key for each customer."
+
+    def test_from_file_detects_block_range_in_multiple_blocks(self):
+        range = JinjaBlock.find_block_range(multiple_blocks, "docs", "potato_name")
+        assert range == (6, 8)
+
+    def test_from_file_extracts_content_in_files_with_multiple_blocks(self):
+        content = JinjaBlock.isolate_content_from_line_range(multiple_blocks, 6, 8)
+        assert content == "The name of the customer's favorite potato dish."

From 4b1029072b8b90c5f566cfca86afa95ff32a7f98 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 16:40:35 -0500
Subject: [PATCH 03/12] fix: Migrate parsing code to use character posisions
 and not line numbers

---
 dbt_meshify/storage/jinja_blocks.py | 15 +++++++++------
 tests/unit/test_jinja_blocks.py     |  8 ++++----
 2 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
index dc78773..c9d2e5d 100644
--- a/dbt_meshify/storage/jinja_blocks.py
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -7,7 +7,7 @@
 @dataclass
 class JinjaBlock:
     """
-    A common data structure for tracking blocks of text that represent Jinja blocks.
+    A data structure for tracking Jinja blocks of text. Includes the start and end character positions, and the content of the block
     """
 
     path: Path
@@ -27,15 +27,15 @@ def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int
             r"{%\s+" + block_type + r"\s+" + name + r"\s+%}", file_content, re.MULTILINE
         ):
             start = match.span()[0]  # .span() gives tuple (start, end)
-            start_line = file_content[:start].count("\n")
+            start_line = start  # file_content[:start].count("\n")
             break
 
         if start_line is None:
             raise Exception(f"Unable to find a {block_type} block with the name {name}.")
 
         for match in re.finditer(r"{%\s+end" + block_type + r"\s+%}", file_content, re.MULTILINE):
-            start = match.span()[0]  # .span() gives tuple (start, end)
-            new_end_line = file_content[:start].count("\n")
+            end = match.span()[1]  # .span() gives tuple (start, end)
+            new_end_line = end  # file_content[:start].count("\n")
 
             if new_end_line >= start_line:
                 end_line = new_end_line
@@ -49,8 +49,11 @@ def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int
     @staticmethod
     def isolate_content_from_line_range(file_content: str, start: int, end: int) -> str:
         """Given content, a start line number, and an end line number, return the content of a Jinja block."""
-        print(file_content.split("\n")[start + 1 :])
-        return "/n".join(file_content.split("\n")[start + 1 : end])
+        raw_content = file_content[start:end]
+        match = re.search(r"{%.*%}\n(.*)\n{%.*%}", raw_content)
+        if match is None:
+            raise Exception("Unable to find the jinja block within the designated range.")
+        return match.group(1)
 
     @classmethod
     def from_file(cls, path: Path, block_type: str, name: str) -> "JinjaBlock":
diff --git a/tests/unit/test_jinja_blocks.py b/tests/unit/test_jinja_blocks.py
index d638458..66f0d5b 100644
--- a/tests/unit/test_jinja_blocks.py
+++ b/tests/unit/test_jinja_blocks.py
@@ -24,16 +24,16 @@
 class TestJinjaBlock:
     def test_from_file_detects_block_range(self):
         range = JinjaBlock.find_block_range(string, "docs", "customer_id")
-        assert range == (2, 4)
+        assert range == (2, 72)
 
     def test_from_file_extracts_content(self):
-        content = JinjaBlock.isolate_content_from_line_range(string, 2, 4)
+        content = JinjaBlock.isolate_content_from_line_range(string, 2, 72)
         assert content == "The unique key for each customer."
 
     def test_from_file_detects_block_range_in_multiple_blocks(self):
         range = JinjaBlock.find_block_range(multiple_blocks, "docs", "potato_name")
-        assert range == (6, 8)
+        assert range == (74, 159)
 
     def test_from_file_extracts_content_in_files_with_multiple_blocks(self):
-        content = JinjaBlock.isolate_content_from_line_range(multiple_blocks, 6, 8)
+        content = JinjaBlock.isolate_content_from_line_range(multiple_blocks, 74, 159)
         assert content == "The name of the customer's favorite potato dish."

From 45fef34ce8367c64a18491e818926baffd6c059c Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 17:01:45 -0500
Subject: [PATCH 04/12] fix: Support formatting characters in block range
 calculations

---
 dbt_meshify/dbt_projects.py         | 11 +++++++++++
 dbt_meshify/storage/jinja_blocks.py | 16 ++++++++--------
 tests/unit/test_jinja_blocks.py     |  9 +++++++--
 3 files changed, 26 insertions(+), 10 deletions(-)

diff --git a/dbt_meshify/dbt_projects.py b/dbt_meshify/dbt_projects.py
index bb22cb9..716443f 100644
--- a/dbt_meshify/dbt_projects.py
+++ b/dbt_meshify/dbt_projects.py
@@ -320,10 +320,21 @@ def find_jinja_blocks(self) -> Dict[str, JinjaBlock]:
         blocks = {}
 
         for unique_id, item in self.manifest.docs.items():
+            if item.package_name != self.name:
+                continue
+
             blocks[unique_id] = JinjaBlock.from_file(
                 path=self.path / item.original_file_path, block_type="docs", name=item.name
             )
 
+        for unique_id, macro in self.manifest.macros.items():
+            if macro.package_name != self.name:
+                continue
+
+            blocks[unique_id] = JinjaBlock.from_file(
+                path=self.path / macro.original_file_path, block_type="macro", name=macro.name
+            )
+
         return blocks
 
     def select_resources(
diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
index c9d2e5d..7585697 100644
--- a/dbt_meshify/storage/jinja_blocks.py
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -24,7 +24,9 @@ def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int
         end_line = None
 
         for match in re.finditer(
-            r"{%\s+" + block_type + r"\s+" + name + r"\s+%}", file_content, re.MULTILINE
+            r"{%-?\s+" + block_type + r"\s+" + name + r"([(a-zA-Z0-9=,_ )]*)\s-?%}",
+            file_content,
+            re.MULTILINE,
         ):
             start = match.span()[0]  # .span() gives tuple (start, end)
             start_line = start  # file_content[:start].count("\n")
@@ -33,7 +35,9 @@ def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int
         if start_line is None:
             raise Exception(f"Unable to find a {block_type} block with the name {name}.")
 
-        for match in re.finditer(r"{%\s+end" + block_type + r"\s+%}", file_content, re.MULTILINE):
+        for match in re.finditer(
+            r"{%-?\s+end" + block_type + r"\s+-?%}", file_content, re.MULTILINE
+        ):
             end = match.span()[1]  # .span() gives tuple (start, end)
             new_end_line = end  # file_content[:start].count("\n")
 
@@ -48,12 +52,8 @@ def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int
 
     @staticmethod
     def isolate_content_from_line_range(file_content: str, start: int, end: int) -> str:
-        """Given content, a start line number, and an end line number, return the content of a Jinja block."""
-        raw_content = file_content[start:end]
-        match = re.search(r"{%.*%}\n(.*)\n{%.*%}", raw_content)
-        if match is None:
-            raise Exception("Unable to find the jinja block within the designated range.")
-        return match.group(1)
+        """Given content, a start position, and an end position, return the content of a Jinja block."""
+        return file_content[start:end]
 
     @classmethod
     def from_file(cls, path: Path, block_type: str, name: str) -> "JinjaBlock":
diff --git a/tests/unit/test_jinja_blocks.py b/tests/unit/test_jinja_blocks.py
index 66f0d5b..020fc0c 100644
--- a/tests/unit/test_jinja_blocks.py
+++ b/tests/unit/test_jinja_blocks.py
@@ -28,7 +28,9 @@ def test_from_file_detects_block_range(self):
 
     def test_from_file_extracts_content(self):
         content = JinjaBlock.isolate_content_from_line_range(string, 2, 72)
-        assert content == "The unique key for each customer."
+        assert (
+            content == "{% docs customer_id %}\nThe unique key for each customer.\n{% enddocs %}"
+        )
 
     def test_from_file_detects_block_range_in_multiple_blocks(self):
         range = JinjaBlock.find_block_range(multiple_blocks, "docs", "potato_name")
@@ -36,4 +38,7 @@ def test_from_file_detects_block_range_in_multiple_blocks(self):
 
     def test_from_file_extracts_content_in_files_with_multiple_blocks(self):
         content = JinjaBlock.isolate_content_from_line_range(multiple_blocks, 74, 159)
-        assert content == "The name of the customer's favorite potato dish."
+        assert (
+            content
+            == "{% docs potato_name %}\nThe name of the customer's favorite potato dish.\n{% enddocs %}"
+        )

From 25c56d6d18389ec72eac0120048b8e9cb89808d1 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 19:36:34 -0500
Subject: [PATCH 05/12] feat: Add file appending. Wire up basic macro appending
 into new projects

---
 dbt_meshify/change.py                       |  2 ++
 dbt_meshify/dbt_projects.py                 |  1 -
 dbt_meshify/main.py                         |  3 ++-
 dbt_meshify/storage/dbt_project_editors.py  | 24 ++++++++++++++++++++-
 dbt_meshify/storage/file_content_editors.py | 12 +++++++++++
 dbt_meshify/storage/file_manager.py         |  6 ++++++
 6 files changed, 45 insertions(+), 3 deletions(-)

diff --git a/dbt_meshify/change.py b/dbt_meshify/change.py
index 429eee8..1871805 100644
--- a/dbt_meshify/change.py
+++ b/dbt_meshify/change.py
@@ -9,6 +9,7 @@ class Operation(str, Enum):
     """An operation describes the type of work being performed."""
 
     Add = "add"
+    Append = "append"
     Update = "update"
     Remove = "remove"
     Copy = "copy"
@@ -17,6 +18,7 @@ class Operation(str, Enum):
 
 prepositions = {
     Operation.Add: "to",
+    Operation.Append: "to",
     Operation.Move: "to",
     Operation.Copy: "to",
     Operation.Update: "in",
diff --git a/dbt_meshify/dbt_projects.py b/dbt_meshify/dbt_projects.py
index 716443f..913320f 100644
--- a/dbt_meshify/dbt_projects.py
+++ b/dbt_meshify/dbt_projects.py
@@ -311,7 +311,6 @@ def __init__(
         resources = self.select_resources(output_key="unique_id")
 
         super().__init__(manifest, project, catalog, name, resources)
-
         self.jinja_blocks: Dict[str, JinjaBlock] = self.find_jinja_blocks()
 
     def find_jinja_blocks(self) -> Dict[str, JinjaBlock]:
diff --git a/dbt_meshify/main.py b/dbt_meshify/main.py
index 9c3001f..ec9248d 100644
--- a/dbt_meshify/main.py
+++ b/dbt_meshify/main.py
@@ -246,7 +246,8 @@ def split(
         change_set = subproject_creator.initialize()
 
         return [change_set]
-    except Exception:
+    except Exception as e:
+        logger.debug(e)
         raise FatalMeshifyException(f"Error creating subproject {subproject.name}")
 
 
diff --git a/dbt_meshify/storage/dbt_project_editors.py b/dbt_meshify/storage/dbt_project_editors.py
index ecea8ba..fa6da1a 100644
--- a/dbt_meshify/storage/dbt_project_editors.py
+++ b/dbt_meshify/storage/dbt_project_editors.py
@@ -22,6 +22,7 @@
 from dbt_meshify.dbt_projects import DbtSubProject
 from dbt_meshify.storage.file_content_editors import NamedList, filter_empty_dict_items
 from dbt_meshify.storage.file_manager import YAMLFileManager, yaml
+from dbt_meshify.storage.jinja_blocks import JinjaBlock
 from dbt_meshify.utilities.contractor import Contractor
 from dbt_meshify.utilities.dependencies import DependenciesUpdater
 from dbt_meshify.utilities.grouper import ResourceGrouper
@@ -185,7 +186,17 @@ def initialize(self) -> ChangeSet:
             elif resource.resource_type in ["macro", "group"]:
                 if hasattr(resource, "patch_path") and resource.patch_path:
                     change_set.add(self.copy_resource_yml(resource))
-                change_set.add(self.copy_resource(resource))
+
+                if resource.unique_id in self.subproject.parent_project.jinja_blocks:
+                    change_set.add(
+                        self.copy_jinja_block(
+                            resource,
+                            self.subproject.parent_project.jinja_blocks[resource.unique_id],
+                        )
+                    )
+
+                else:
+                    change_set.add(self.copy_resource(resource))
 
             else:
                 logger.debug(
@@ -237,6 +248,17 @@ def move_resource(self, resource: Resource) -> FileChange:
             source=self.subproject.parent_project.resolve_file_path(resource),
         )
 
+    def copy_jinja_block(self, resource: Resource, jinja_block: JinjaBlock) -> FileChange:
+        """Move an existing jinja block to a new project"""
+
+        return FileChange(
+            operation=Operation.Append,
+            entity_type=EntityType.Code,
+            identifier=resource.name,
+            path=self.subproject.resolve_file_path(resource),
+            data=jinja_block.content,
+        )
+
     def copy_resource(self, resource: Resource) -> FileChange:
         """
         Copy a resource file from one project to another
diff --git a/dbt_meshify/storage/file_content_editors.py b/dbt_meshify/storage/file_content_editors.py
index ecd9997..5f88c44 100644
--- a/dbt_meshify/storage/file_content_editors.py
+++ b/dbt_meshify/storage/file_content_editors.py
@@ -107,6 +107,18 @@ def add(change: FileChange):
 
         RawFileManager.write_file(path=change.path, content=change.data)
 
+    @staticmethod
+    def append(change: FileChange):
+        """Append data to an existing file."""
+
+        if not change.path.parent.exists():
+            change.path.parent.mkdir(parents=True, exist_ok=True)
+
+        if change.data is None:
+            return RawFileManager.touch_file(change.path)
+
+        RawFileManager.append_file(path=change.path, content=change.data)
+
     @staticmethod
     def update(change: FileChange):
         """Update data to a new file."""
diff --git a/dbt_meshify/storage/file_manager.py b/dbt_meshify/storage/file_manager.py
index 76a1e72..e845dcd 100644
--- a/dbt_meshify/storage/file_manager.py
+++ b/dbt_meshify/storage/file_manager.py
@@ -62,6 +62,12 @@ def write_file(path: Path, content: str) -> None:
         """Write a string value to a file in the filesystem"""
         path.write_text(content)
 
+    @staticmethod
+    def append_file(path: Path, content: str) -> None:
+        """Append a string value to a file in the filesystem"""
+        with open(path, "a") as file:
+            file.write(content)
+
     @staticmethod
     def copy_file(source_path: Path, target_path: Path) -> None:
         if not target_path.parent.exists():

From 8a0fd7b3ca775caf7be568579afdf514da3a4b17 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 19:44:51 -0500
Subject: [PATCH 06/12] docs: Add some TODOs to keep me honest

---
 dbt_meshify/storage/dbt_project_editors.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/dbt_meshify/storage/dbt_project_editors.py b/dbt_meshify/storage/dbt_project_editors.py
index fa6da1a..3c5840b 100644
--- a/dbt_meshify/storage/dbt_project_editors.py
+++ b/dbt_meshify/storage/dbt_project_editors.py
@@ -188,6 +188,9 @@ def initialize(self) -> ChangeSet:
                     change_set.add(self.copy_resource_yml(resource))
 
                 if resource.unique_id in self.subproject.parent_project.jinja_blocks:
+                    # TODO: How do we know when to copy vs move?
+                    # TODO: What about docs? Docs have no graph, right?
+
                     change_set.add(
                         self.copy_jinja_block(
                             resource,

From ee6ee20941eb975ce8a4cf296a49a0f80b586838 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 21:07:27 -0500
Subject: [PATCH 07/12] feat: Add doc reference tracking and support writing
 docs to destination projects

---
 dbt_meshify/dbt_projects.py                | 37 +++++++++++++++++++++-
 dbt_meshify/main.py                        |  2 +-
 dbt_meshify/storage/dbt_project_editors.py |  9 ++++--
 dbt_meshify/storage/jinja_blocks.py        |  9 +++++-
 4 files changed, 52 insertions(+), 5 deletions(-)

diff --git a/dbt_meshify/dbt_projects.py b/dbt_meshify/dbt_projects.py
index 913320f..1ba0930 100644
--- a/dbt_meshify/dbt_projects.py
+++ b/dbt_meshify/dbt_projects.py
@@ -26,7 +26,7 @@
 
 from dbt_meshify.dbt import Dbt
 from dbt_meshify.exceptions import FatalMeshifyException
-from dbt_meshify.storage.jinja_blocks import JinjaBlock
+from dbt_meshify.storage.jinja_blocks import JinjaBlock, find_doc_reference
 
 
 class BaseDbtProject:
@@ -421,6 +421,8 @@ def __init__(
         self.groups = self._get_indirect_groups()
         self._rename_project()
 
+        self._referenced_docs: Optional[Set[str]] = None
+
     def _rename_project(self) -> None:
         """
         edits the project yml to take any instance of the parent project name and update it to the subproject name
@@ -457,6 +459,39 @@ def _get_custom_macros(self) -> Set[str]:
                 macros_set.update(self._get_macro_dependencies(macro))
         return macros_set
 
+    @property
+    def referenced_docs(self) -> Set[str]:
+        """Return a list of all docs referenced within this SubProject."""
+
+        if self._referenced_docs:
+            return self._referenced_docs
+
+        docs = set()
+        for unique_id in self.resources:
+            if unique_id.startswith("test."):
+                continue
+
+            node = self.get_manifest_node(unique_id)
+
+            if node is None:
+                continue
+
+            if hasattr(node, "raw_code"):
+                docs.update(find_doc_reference(node.raw_code))
+
+            if hasattr(node, "patch_path"):
+                path = self.parent_project.resolve_patch_path(node)
+                if path.exists():
+                    with open(path) as file:
+                        docs.update(find_doc_reference(file.read()))
+
+        # Use the search name for the doc to resolve a unique_id for the doc resource.
+        self._referenced_docs = {
+            unique_id for unique_id, doc in self.manifest.docs.items() if doc.name in docs
+        }
+
+        return self._referenced_docs
+
     def _get_indirect_groups(self) -> Set[str]:
         """
         get a set of group unique_ids for all the selected resources
diff --git a/dbt_meshify/main.py b/dbt_meshify/main.py
index ec9248d..3742e43 100644
--- a/dbt_meshify/main.py
+++ b/dbt_meshify/main.py
@@ -247,7 +247,7 @@ def split(
 
         return [change_set]
     except Exception as e:
-        logger.debug(e)
+        logger.exception(e)  # TODO: Remove this line!
         raise FatalMeshifyException(f"Error creating subproject {subproject.name}")
 
 
diff --git a/dbt_meshify/storage/dbt_project_editors.py b/dbt_meshify/storage/dbt_project_editors.py
index 3c5840b..423bce8 100644
--- a/dbt_meshify/storage/dbt_project_editors.py
+++ b/dbt_meshify/storage/dbt_project_editors.py
@@ -142,7 +142,12 @@ def initialize(self) -> ChangeSet:
             f"Identifying operations required to split {subproject.name} from {subproject.parent_project.name}."
         )
 
-        for unique_id in subproject.resources | subproject.custom_macros | subproject.groups:
+        for unique_id in (
+            subproject.resources
+            | subproject.custom_macros
+            | subproject.groups
+            | subproject.referenced_docs
+        ):
             resource = subproject.get_manifest_node(unique_id)
             if not resource:
                 raise KeyError(f"Resource {unique_id} not found in manifest")
@@ -183,7 +188,7 @@ def initialize(self) -> ChangeSet:
                 ):
                     change_set.extend(reference_updater.update_parent_refs(resource))
 
-            elif resource.resource_type in ["macro", "group"]:
+            elif resource.resource_type in ["macro", "group", "doc"]:
                 if hasattr(resource, "patch_path") and resource.patch_path:
                     change_set.add(self.copy_resource_yml(resource))
 
diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
index 7585697..ddf7b17 100644
--- a/dbt_meshify/storage/jinja_blocks.py
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -1,7 +1,7 @@
 import re
 from dataclasses import dataclass
 from pathlib import Path
-from typing import Tuple
+from typing import Set, Tuple
 
 
 @dataclass
@@ -68,3 +68,10 @@ def from_file(cls, path: Path, block_type: str, name: str) -> "JinjaBlock":
         return cls(
             path=path, block_type=block_type, name=name, start=start, end=end, content=content
         )
+
+
+def find_doc_reference(content: str) -> Set[str]:
+    """Find all doc block references within a string."""
+    matches = re.findall(r"{{\sdoc\(\'?\"?([a-zA-Z0-9_\-\.]+)\'?\"?\)\s}}", content)
+
+    return set(matches)

From 688cebbecd4a40e667f762465e4ef8ab641a161b Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 21:25:04 -0500
Subject: [PATCH 08/12] tests: Tweak tests to cover new macro and doc copying
 behavior

---
 .../split/split_proj/macros/_macros.yml          |  4 +++-
 .../split/split_proj/macros/cents_to_dollars.sql |  5 +++++
 tests/integration/test_split_command.py          | 16 ++++++++++++++++
 3 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/test-projects/split/split_proj/macros/_macros.yml b/test-projects/split/split_proj/macros/_macros.yml
index 40d3dab..1767960 100644
--- a/test-projects/split/split_proj/macros/_macros.yml
+++ b/test-projects/split/split_proj/macros/_macros.yml
@@ -1,3 +1,5 @@
 macros:
   - name: cents_to_dollars
-    description: Converts cents to dollars
\ No newline at end of file
+    description: Converts cents to dollars
+  - name: dollars_to_cents
+    description: Converts dollars to cents
diff --git a/test-projects/split/split_proj/macros/cents_to_dollars.sql b/test-projects/split/split_proj/macros/cents_to_dollars.sql
index efe3361..a48174a 100644
--- a/test-projects/split/split_proj/macros/cents_to_dollars.sql
+++ b/test-projects/split/split_proj/macros/cents_to_dollars.sql
@@ -3,3 +3,8 @@
 {% macro cents_to_dollars(column_name, precision=2) -%}
     ({{ column_name }} / 100)::{{ type_numeric() }}(16, {{ precision }})
 {%- endmacro %}
+
+
+{% macro dollars_to_cents(column_name) -%}
+    ({{ column_name }} * 100)::{{ type_numeric() }}(16, 0)
+{%- endmacro %}
diff --git a/tests/integration/test_split_command.py b/tests/integration/test_split_command.py
index bd15751..6801624 100644
--- a/tests/integration/test_split_command.py
+++ b/tests/integration/test_split_command.py
@@ -43,6 +43,13 @@ def test_split_one_model(self, project):
         ).read_text()
         assert x_proj_ref in child_sql
 
+        # Copied a referenced docs block
+        assert (Path(dest_project_path) / "my_new_project" / "models" / "docs.md").exists()
+        assert (
+            "customer_id"
+            in (Path(dest_project_path) / "my_new_project" / "models" / "docs.md").read_text()
+        )
+
     def test_split_one_model_one_source(self, project):
         runner = CliRunner()
         result = runner.invoke(
@@ -111,6 +118,15 @@ def test_split_one_model_one_source_custom_macro(self, project):
         assert (
             Path(dest_project_path) / "my_new_project" / "macros" / "cents_to_dollars.sql"
         ).exists()
+
+        # Confirm that we did not bring over an unreferenced dollars_to_cents macro
+        assert (
+            "dollars_to_cents"
+            not in (
+                Path(dest_project_path) / "my_new_project" / "macros" / "cents_to_dollars.sql"
+            ).read_text()
+        )
+
         # copied custom macro parents too!
         assert (
             Path(dest_project_path) / "my_new_project" / "macros" / "type_numeric.sql"

From 1d9008fceede0520d2d83e8ee6f20c367b6830cd Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 21:33:10 -0500
Subject: [PATCH 09/12] style: Remove unnecessary todos

---
 dbt_meshify/storage/dbt_project_editors.py | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/dbt_meshify/storage/dbt_project_editors.py b/dbt_meshify/storage/dbt_project_editors.py
index 423bce8..2b45c91 100644
--- a/dbt_meshify/storage/dbt_project_editors.py
+++ b/dbt_meshify/storage/dbt_project_editors.py
@@ -193,9 +193,6 @@ def initialize(self) -> ChangeSet:
                     change_set.add(self.copy_resource_yml(resource))
 
                 if resource.unique_id in self.subproject.parent_project.jinja_blocks:
-                    # TODO: How do we know when to copy vs move?
-                    # TODO: What about docs? Docs have no graph, right?
-
                     change_set.add(
                         self.copy_jinja_block(
                             resource,

From 892dd650c256a512cf064ac5e55cf33c646bc45d Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Thu, 28 Dec 2023 21:35:20 -0500
Subject: [PATCH 10/12] refactor: Rename a method in JinjaBlock

---
 dbt_meshify/storage/jinja_blocks.py | 6 ++----
 tests/unit/test_jinja_blocks.py     | 4 ++--
 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
index ddf7b17..dcc4630 100644
--- a/dbt_meshify/storage/jinja_blocks.py
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -51,7 +51,7 @@ def find_block_range(file_content: str, block_type: str, name: str) -> Tuple[int
         return start_line, end_line
 
     @staticmethod
-    def isolate_content_from_line_range(file_content: str, start: int, end: int) -> str:
+    def isolate_content(file_content: str, start: int, end: int) -> str:
         """Given content, a start position, and an end position, return the content of a Jinja block."""
         return file_content[start:end]
 
@@ -61,9 +61,7 @@ def from_file(cls, path: Path, block_type: str, name: str) -> "JinjaBlock":
 
         file_content = open(path).read()
         start, end = cls.find_block_range(file_content, block_type, name)
-        content = cls.isolate_content_from_line_range(
-            file_content=file_content, start=start, end=end
-        )
+        content = cls.isolate_content(file_content=file_content, start=start, end=end)
 
         return cls(
             path=path, block_type=block_type, name=name, start=start, end=end, content=content
diff --git a/tests/unit/test_jinja_blocks.py b/tests/unit/test_jinja_blocks.py
index 020fc0c..6a8e896 100644
--- a/tests/unit/test_jinja_blocks.py
+++ b/tests/unit/test_jinja_blocks.py
@@ -27,7 +27,7 @@ def test_from_file_detects_block_range(self):
         assert range == (2, 72)
 
     def test_from_file_extracts_content(self):
-        content = JinjaBlock.isolate_content_from_line_range(string, 2, 72)
+        content = JinjaBlock.isolate_content(string, 2, 72)
         assert (
             content == "{% docs customer_id %}\nThe unique key for each customer.\n{% enddocs %}"
         )
@@ -37,7 +37,7 @@ def test_from_file_detects_block_range_in_multiple_blocks(self):
         assert range == (74, 159)
 
     def test_from_file_extracts_content_in_files_with_multiple_blocks(self):
-        content = JinjaBlock.isolate_content_from_line_range(multiple_blocks, 74, 159)
+        content = JinjaBlock.isolate_content(multiple_blocks, 74, 159)
         assert (
             content
             == "{% docs potato_name %}\nThe name of the customer's favorite potato dish.\n{% enddocs %}"

From 08246a598ad5472a56a8b5ef11aaaa34a0b6fae0 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Tue, 2 Jan 2024 13:44:36 -0500
Subject: [PATCH 11/12] fix: Convert from open().read() to read_text()

---
 dbt_meshify/storage/jinja_blocks.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dbt_meshify/storage/jinja_blocks.py b/dbt_meshify/storage/jinja_blocks.py
index dcc4630..795f061 100644
--- a/dbt_meshify/storage/jinja_blocks.py
+++ b/dbt_meshify/storage/jinja_blocks.py
@@ -59,7 +59,7 @@ def isolate_content(file_content: str, start: int, end: int) -> str:
     def from_file(cls, path: Path, block_type: str, name: str) -> "JinjaBlock":
         """Find a specific Jinja block within a file, based on the block type and the name."""
 
-        file_content = open(path).read()
+        file_content = path.read_text()
         start, end = cls.find_block_range(file_content, block_type, name)
         content = cls.isolate_content(file_content=file_content, start=start, end=end)
 

From 0163c70ffc190aa6f720f1dd8e29ded6680f3aa2 Mon Sep 17 00:00:00 2001
From: Nicholas Yager <yager@nicholasyager.com>
Date: Tue, 2 Jan 2024 13:49:59 -0500
Subject: [PATCH 12/12] fix: Raise an exceptioj in referenced_docs if a node
 cannot be found

---
 dbt_meshify/dbt_projects.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/dbt_meshify/dbt_projects.py b/dbt_meshify/dbt_projects.py
index 1ba0930..a0ed37f 100644
--- a/dbt_meshify/dbt_projects.py
+++ b/dbt_meshify/dbt_projects.py
@@ -474,7 +474,7 @@ def referenced_docs(self) -> Set[str]:
             node = self.get_manifest_node(unique_id)
 
             if node is None:
-                continue
+                raise Exception(f"Unable to find referenced node {node}")
 
             if hasattr(node, "raw_code"):
                 docs.update(find_doc_reference(node.raw_code))