Semantic Diff for SQL
Motivation
Software is constantly changing and evolving, and identifying what has changed and reviewing those changes is an integral part of the development process. SQL code is no exception to this.
Text-based diff tools such as git diff
, when applied to a code base, have certain limitations. First, they can only detect insertions and deletions, not movements or updates of individual pieces of code. Second, such tools can only detect changes between lines of text, which is too coarse for something as granular and detailed as source code. Additionally, the outcome of such a diff is dependent on the underlying code formatting, and yields different results if the formatting should change.
Consider the following diff generated by Git:
Semantically the query hasn’t changed. The two arguments b
and c
have been swapped (moved), posing no impact on the output of the query. Yet Git replaced the whole affected expression alongside a bulk of unrelated elements.
The alternative to text-based diffing is to compare Abstract Syntax Trees (AST) instead. The main advantage of ASTs are that they are a direct product of code parsing, which represents the underlying code structure at any desired level of granularity. Comparing ASTs may yield extremely precise diffs; changes such as code movements and updates can also be detected. Even more importantly, this approach facilitates additional use cases beyond eyeballing two versions of source code side by side.
The use cases I had in mind for SQL when I decided to embark on this journey of semantic diffing were the following:
- Query similarity score. Identifying which parts the two queries have in common to automatically suggest opportunities for consolidation, creation of intermediate/staging tables, and so on.
- Differentiating between cosmetic / structural changes and functional ones. For example when a nested query is refactored into a common table expression (CTE), this kind of change doesn’t have any functional impact on either a query or its outcome.
- Automatic suggestions about the need to retroactively backfill data. This is especially important for pipelines that populate very large tables for which restatement is a runtime-intensive procedure. The ability to discern between simple code movements and actual modifications can help assess the impact of a change and make suggestions accordingly.
The implementation discussed in this post is now a part of the SQLGlot library. You can find a complete source code in the diff.py module. The choice of SQLglot was an obvious one due to its simple but powerful API, lack of external dependencies and, more importantly, extensive list of supported SQL dialects.
The Search for a Solution
When it comes to any diffing tool (not just a semantic one), the primary challenge is to match as many elements of compared entities as possible. Once such a set of matching elements is available, deriving a sequence of changes becomes an easy task.
If our elements have unique identifiers associated with them (for example, an element’s ID in DOM), the matching problem is trivial. However, the SQL syntax trees that we are comparing have neither unique keys nor object identifiers that can be used for the purposes of matching. So, how do we suppose to find pairs of nodes that are related?
To better illustrate the problem, consider comparing the following SQL expressions: SELECT a + b + c, d, e
and SELECT a - b + c, e, f
. Matching individual nodes from respective syntax trees can be visualized as follows:
Figure 1: Example of node matching for two SQL expression trees.
By looking at the figure of node matching for two SQL expression trees above, we conclude that the following changes should be captured by our solution:
- Inserted nodes:
Sub
andf
. These are the nodes from the target AST which do not have a matching node in the source AST. - Removed nodes:
Add
andd
. These are the nodes from the source AST which do not have a counterpart in the target AST. - Remaining nodes must be identified as unchanged.
It should be clear at this point that if we manage to match nodes in the source tree with their counterparts in the target tree, then computing the diff becomes a trivial matter.
Naïve Brute-Force
The naïve solution would be to try all different permutations of node pair combinations, and see which set of pairs performs the best based on some type of heuristics. The runtime cost of such a solution quickly reaches the escape velocity; if both trees had only 10 nodes each, the number of such sets would approximately be 10! ^ 2 = 3.6M ^ 2 ~= 13 * 10^12. This is a very bad case of factorial complexity (to be precise, it’s actually much worse - O(n! ^ 2) - but I couldn’t come up with a name for it), so there is little need to explore this approach any further.
Myers Algorithm
After the naïve approach was proven to be infeasible, the next question I asked myself was “how does git diff work?”. This question led me to discover the Myers diff algorithm [1]. This algorithm has been designed to compare sequences of strings. At its core, it’s looking for the shortest path on a graph of possible edits that transform the first sequence into the second one, while heavily rewarding those paths that lead to longest subsequences of unchanged elements. There’s a lot of material out there describing this algorithm in greater detail. I found James Coglan’s series of blog posts to be the most comprehensive.
Therefore, I had this “brilliant” (actually not) idea to transform trees into sequences by traversing them in topological order, and then applying the Myers algorithm on resulting sequences while using a custom heuristics when checking the equality of two nodes. Unsurprisingly, comparing sequences of strings is quite different from comparing hierarchical tree structures, and by flattening trees into sequences, we lose a lot of relevant context. This resulted in a terrible performance of this algorithm on ASTs. It often matched completely unrelated nodes, even when the two trees were mostly the same, and produced extremely inaccurate lists of changes overall. After playing around with it a little and tweaking my equality heuristics to improve accuracy, I ultimately scrapped the whole implementation and went back to the drawing board.
Change Distiller
The algorithm I settled on at the end was Change Distiller, created by Fluri et al. [2], which in turn is an improvement over the core idea described by Chawathe et al. [3].
The algorithm consists of two high-level steps:
- Finding appropriate matchings between pairs of nodes that are part of compared ASTs. Identifying what is meant by “appropriate” matching is also a part of this step.
- Generating the so-called “edit script” from the matching set built in the 1st step. The edit script is a sequence of edit operations (for example, insert, remove, update, etc.) on individual tree nodes, such that when applied as transformations on the source AST, it eventually becomes the target AST. In general, the shorter the sequence, the better. The length of the edit script can be used to compare the performance of different algorithms, though this is not the only metric that matters.
The rest of this section is dedicated to the Python implementation of the steps above using the AST implementation provided by the SQLGlot library.
Building the Matching Set
Matching Leaves
We begin composing the matching set by matching the leaf nodes. Leaf nodes are the nodes that do not have any children nodes (such as literals, identifiers, etc.). In order to match them, we gather all the leaf nodes from the source tree and generate a cartesian product with all the leaves from the target tree, while comparing pairs created this way and assigning them a similarity score. During this stage, we also exclude pairs that don’t pass basic matching criteria. Then, we pick pairs that scored the highest while making sure that each node is matched no more than once.
Using the example provided at the beginning of the post, the process of building an initial set of candidate matchings can be seen on Figure 2.
Figure 2: Building a set of candidate matchings between leaf nodes. The third item in each triplet represents a similarity score between two nodes.
First, let’s analyze the similarity score. Then, we’ll discuss matching criteria.
The similarity score proposed by Fluri et al. [2] is a dice coefficient applied to bigrams of respective node values. A bigram is a sequence of two adjacent elements from a string computed in a sliding window fashion:
def bigram(string):
count = max(0, len(string) - 1)
return [string[i : i + 2] for i in range(count)]
For reasons that will become clear shortly, we actually need to compute bigram histograms rather than just sequences:
from collections import defaultdict
def bigram_histo(string):
count = max(0, len(string) - 1)
bigram_histo = defaultdict(int)
for i in range(count):
bigram_histo[string[i : i + 2]] += 1
return bigram_histo
The dice coefficient formula looks like following:
Where X is a bigram of the source node and Y is a bigram of the second one. What this essentially does is count the number of bigram elements the two nodes have in common, multiply it by 2, and then divide by the total number of elements in both bigrams. This is where bigram histograms come in handy:
def dice_coefficient(source, target):
source_histo = bigram_histo(source.sql())
target_histo = bigram_histo(target.sql())
total_grams = (
sum(source_histo.values()) + sum(target_histo.values())
)
if not total_grams:
return 1.0 if source == target else 0.0
overlap_len = 0
overlapping_grams = set(source_histo) & set(target_histo)
for g in overlapping_grams:
overlap_len += min(source_histo[g], target_histo[g])
return 2 * overlap_len / total_grams
To compute a bigram given a tree node, we first transform the node into its canonical SQL representation,so that the Literal(123)
node becomes just “123” and the Identifier(“a”)
node becomes just “a”. We also handle a scenario when strings are too short to derive bigrams. In this case, we fallback to checking the two nodes for equality.
Now when we know how to compute the similarity score, we can take care of the matching criteria for leaf nodes. In the original paper [2], the matching criteria is formalized as follows:
The two nodes are matched if two conditions are met:
- The node labels match (in our case labels are just node types).
- The similarity score for node values is greater than or equal to some threshold “f”. The authors of the paper recommend setting the value of “f” to 0.6.
With building blocks in place, we can now build a matching set for leaf nodes. First, we generate a list of candidates for matching:
from heapq import heappush, heappop
candidate_matchings = []
source_leaves = _get_leaves(self._source)
target_leaves = _get_leaves(self._target)
for source_leaf in source_leaves:
for target_leaf in target_leaves:
if _is_same_type(source_leaf, target_leaf):
similarity_score = dice_coefficient(
source_leaf, target_leaf
)
if similarity_score >= 0.6:
heappush(
candidate_matchings,
(
-similarity_score,
len(candidate_matchings),
source_leaf,
target_leaf,
),
)
In the implementation above, we push each matching pair onto the heap to automatically maintain the correct order based on the assigned similarity score.
Finally, we build the initial matching set by picking leaf pairs with the highest score:
matching_set = set()
while candidate_matchings:
_, _, source_leaf, target_leaf = heappop(candidate_matchings)
if (
source_leaf in unmatched_source_nodes
and target_leaf in unmatched_target_nodes
):
matching_set.add((source_leaf, target_leaf))
unmatched_source_nodes.remove(source_leaf)
unmatched_target_nodes.remove(target_leaf)
To finalize the matching set, we should now proceed with matching inner nodes.
Matching Inner Nodes
Matching inner nodes is quite similar to matching leaf nodes, with the following two distinctions:
- Rather than ranking a set of possible candidates, we pick the first node pair that passes the matching criteria.
- The matching criteria itself has been extended to account for the number of leaf nodes the pair of inner nodes have in common.
Figure 3: Matching inner nodes based on their type as well as how many of their leaf nodes have been previously matched.
Let’s start with the matching criteria. The criteria is formalized as follows:
Alongside already familiar similarity score and node type criteria, there is a new one in the middle: the ratio of leaf nodes that the two nodes have in common must exceed some threshold “t”. The recommended value for “t” is also 0.6. Counting the number of common leaf nodes is pretty straightforward, since we already have the complete matching set for leaves. All we need to do is count how many matching pairs do leaf nodes from the two compared inner nodes form.
There are two additional heuristics associated with this matching criteria:
- Inner node similarity weighting: if the similarity score between the node values doesn’t pass the threshold “f” but the ratio of common leaf nodes (“t”) is greater than or equal to 0.8, then the matching is considered successful.
- The threshold “t” is reduced to 0.4 for inner nodes with the number of leaf nodes equal to 4 or less, in order to decrease the false negative rate for small subtrees.
We now only have to iterate through the remaining unmatched nodes and form matching pairs based on the outlined criteria:
leaves_matching_set = matching_set.copy()
for source_node in unmatched_source_nodes.copy():
for target_node in unmatched_target_nodes:
if _is_same_type(source_node, target_node):
source_leaves = set(_get_leaves(source_node))
target_leaves = set(_get_leaves(target_node))
max_leaves_num = max(len(source_leaves), len(target_leaves))
if max_leaves_num:
common_leaves_num = sum(
1 if s in source_leaves and t in target_leaves else 0
for s, t in leaves_matching_set
)
leaf_similarity_score = common_leaves_num / max_leaves_num
else:
leaf_similarity_score = 0.0
adjusted_t = (
0.6
if min(len(source_leaves), len(target_leaves)) > 4
else 0.4
)
if leaf_similarity_score >= 0.8 or (
leaf_similarity_score >= adjusted_t
and dice_coefficient(source_node, target_node) >= 0.6
):
matching_set.add((source_node, target_node))
unmatched_source_nodes.remove(source_node)
unmatched_target_nodes.remove(target_node)
break
After the matching set is formed, we can proceed with generation of the edit script, which will be the algorithm’s output.
Generating the Edit Script
At this point, we should have the following 3 sets at our disposal:
- The set of matched node pairs.
- The set of remaining unmatched nodes from the source tree.
- The set of remaining unmatched nodes from the target tree.
We can derive 3 kinds of edits from the matching set: either the node’s value was updated (Update), the node was moved to a different position within the tree (Move), or the node remained unchanged (Keep). Note that the Move case is not mutually exclusive with the other two. The node could have been updated or could have remained the same while at the same time its position within its parent node or the parent node itself could have changed. All unmatched nodes from the source tree are the ones that were removed (Remove), while unmatched nodes from the target tree are the ones that were inserted (Insert).
The latter two cases are pretty straightforward to implement:
edit_script = []
for removed_node in unmatched_source_nodes:
edit_script.append(Remove(removed_node))
for inserted_node in unmatched_target_nodes:
edit_script.append(Insert(inserted_node))
Traversing the matching set requires a little more thought:
for source_node, target_node in matching_set:
if (
not isinstance(source_node, LEAF_EXPRESSION_TYPES)
or source_node == target_node
):
move_edits = generate_move_edits(
source_node, target_node, matching_set
)
edit_script.extend(move_edits)
edit_script.append(Keep(source_node, target_node))
else:
edit_script.append(Update(source_node, target_node))
If a matching pair represents a pair of leaf nodes, we check if they are the same to decide whether an update took place. For inner node pairs, we also need to compare the positions of their respective children to detect node movements. Chawathe et al. [3] suggest applying the longest common subsequence (LCS) algorithm which, no surprise here, was described by Myers himself [1]. There is a small catch, however: instead of checking the equality of two children nodes, we need to check whether the two nodes form a pair that is a part of our matching set.
Now with this knowledge, the implementation becomes straightforward:
def generate_move_edits(source, target, matching_set):
source_children = _get_child_nodes(source)
target_children = _get_child_nodes(target)
lcs = set(
_longest_common_subsequence(
source_children,
target_children,
lambda l, r: (l, r) in matching_set
)
)
move_edits = []
for node in source_children:
if node not in lcs and node not in unmatched_source_nodes:
move_edits.append(Move(node))
return move_edits
I left out the implementation of the LCS algorithm itself here, but there are plenty of implementation choices out there that can be easily looked up.
Output
The implemented algorithm produces the output that resembles the following:
>>> from sqlglot import parse_one, diff
>>> diff(parse_one("SELECT a + b + c, d, e"), parse_one("SELECT a - b + c, e, f"))
Remove(Add)
Remove(Column(d))
Remove(Identifier(d))
Insert(Sub)
Insert(Column(f))
Insert(Identifier(f))
Keep(Select, Select)
Keep(Add, Add)
Keep(Column(a), Column(a))
Keep(Identifier(a), Identifier(a))
Keep(Column(b), Column(b))
Keep(Identifier(b), Identifier(b))
Keep(Column(c), Column(c))
Keep(Identifier(c), Identifier(c))
Keep(Column(e), Column(e))
Keep(Identifier(e), Identifier(e))
Note that the output above is abbreviated. The string representation of actual AST nodes is significantly more verbose.
The implementation works especially well when coupled with the SQLGlot’s query optimizer which can be used to produce canonical representations of compared queries:
>>> schema={"t": {"a": "INT", "b": "INT", "c": "INT", "d": "INT"}}
>>> source = """
... SELECT 1 + 1 + a
... FROM t
... WHERE b = 1 OR (c = 2 AND d = 3)
... """
>>> target = """
... SELECT 2 + a
... FROM t
... WHERE (b = 1 OR c = 2) AND (b = 1 OR d = 3)
... """
>>> optimized_source = optimize(parse_one(source), schema=schema)
>>> optimized_target = optimize(parse_one(target), schema=schema)
>>> edit_script = diff(optimized_source, optimized_target)
>>> sum(0 if isinstance(e, Keep) else 1 for e in edit_script)
0
Optimizations
The worst case runtime complexity of this algorithm is not exactly stellar: O(n^2 * log n^2). This is because of the leaf matching process, which involves ranking a cartesian product between all leaf nodes of compared trees. Unsurprisingly, the algorithm takes a considerable time to finish for bigger queries.
There are still a few basic things we can do in our implementation to help improve performance:
- Refer to individual node objects using their identifiers (Python’s id()) instead of direct references in sets. This helps avoid costly recursive hash calculations and equality checks.
- Cache bigram histograms to avoid computing them more than once for the same node.
- Compute the canonical SQL string representation for each tree once while caching string representations of all inner nodes. This prevents redundant tree traversals when bigrams are computed.
At the time of writing only the first two optimizations have been implemented, so there is an opportunity to contribute for anyone who’s interested.
Alternative Solutions
This section is dedicated to solutions that I’ve investigated, but haven’t tried.
First, this section wouldn’t be complete without Tristan Hume’s blog post. Tristan’s solution has a lot in common with the Myers algorithm plus heuristics that is much more clever than what I came up with. The implementation relies on a combination of dynamic programming and A* search algorithm to explore the space of possible matchings and pick the best ones. It seemed to have worked well for Tistan’s specific use case, but after my negative experience with the Myers algorithm, I decided to try something different.
Another notable approach is the Gumtree algorithm by Falleri et al. [4]. I discovered this paper after I’d already implemented the algorithm that is the main focus of this post. In sections 5.2 and 5.3 of their paper, the authors compare the two algorithms side by side and claim that Gumtree is significantly better in terms of both runtime performance and accuracy when evaluated on 12 792 pairs of Java source files. This doesn’t surprise me, as the algorithm takes the height of subtrees into account. In my tests, I definitely saw scenarios in which this context would have helped. On top of that, the authors promise O(n^2) runtime complexity in the worst case which, given the Change Distiller's O(n^2 * log n^2), looks particularly tempting. I hope to try this algorithm out at some point, and there is a good chance you see me writing about it in my future posts.
Conclusion
The Change Distiller algorithm yielded quite satisfactory results in most of my tests. The scenarios in which it fell short mostly concerned identical (or very similar) subtrees located in different parts of the AST. In those cases, node mismatches were frequent and, as a result, edit scripts were somewhat suboptimal.
Additionally, the runtime performance of the algorithm leaves a lot to be desired. On trees with 1000 leaf nodes each, the algorithm takes a little under 2 seconds to complete. My implementation still has room for improvement, but this should give you a rough idea of what to expect. It appears that the Gumtree algorithm [4] can help address both of these points. I hope to find bandwidth to work on it soon and then compare the two algorithms side-by-side to find out which one performs better on SQL specifically. In the meantime, Change Distiller definitely gets the job done, and I can now proceed with applying it to some of the use cases I mentioned at the beginning of this post.
I’m also curious to learn whether other folks in the industry faced a similar problem, and how they approached it. If you did something similar, I’m interested to hear about your experience.
References
[1] Eugene W. Myers. An O(ND) Difference Algorithm and Its Variations. Algorithmica 1(2): 251-266 (1986)
[2] B. Fluri, M. Wursch, M. Pinzger, and H. Gall. Change Distilling: Tree differencing for fine-grained source code change extraction. IEEE Trans. Software Eng., 33(11):725–743, 2007.
[3] S.S. Chawathe, A. Rajaraman, H. Garcia-Molina, and J. Widom. Change Detection in Hierarchically Structured Information. Proc. ACM Sigmod Int’l Conf. Management of Data, pp. 493-504, June 1996
[4] Jean-Rémy Falleri, Floréal Morandat, Xavier Blanc, Matias Martinez, Martin Monperrus. Fine-grained and Accurate Source Code Differencing. Proceedings of the International Conference on Automated Software Engineering, 2014, Västeras, Sweden. pp.313-324, 10.1145/2642937.2642982. hal-01054552
1""" 2.. include:: ../posts/sql_diff.md 3 4---- 5""" 6 7from __future__ import annotations 8 9import typing as t 10from collections import defaultdict 11from dataclasses import dataclass 12from heapq import heappop, heappush 13 14from sqlglot import Dialect, expressions as exp 15from sqlglot.helper import ensure_list 16 17 18@dataclass(frozen=True) 19class Insert: 20 """Indicates that a new node has been inserted""" 21 22 expression: exp.Expression 23 24 25@dataclass(frozen=True) 26class Remove: 27 """Indicates that an existing node has been removed""" 28 29 expression: exp.Expression 30 31 32@dataclass(frozen=True) 33class Move: 34 """Indicates that an existing node's position within the tree has changed""" 35 36 expression: exp.Expression 37 38 39@dataclass(frozen=True) 40class Update: 41 """Indicates that an existing node has been updated""" 42 43 source: exp.Expression 44 target: exp.Expression 45 46 47@dataclass(frozen=True) 48class Keep: 49 """Indicates that an existing node hasn't been changed""" 50 51 source: exp.Expression 52 target: exp.Expression 53 54 55if t.TYPE_CHECKING: 56 from sqlglot._typing import T 57 58 Edit = t.Union[Insert, Remove, Move, Update, Keep] 59 60 61def diff( 62 source: exp.Expression, 63 target: exp.Expression, 64 matchings: t.List[t.Tuple[exp.Expression, exp.Expression]] | None = None, 65 delta_only: bool = False, 66 **kwargs: t.Any, 67) -> t.List[Edit]: 68 """ 69 Returns the list of changes between the source and the target expressions. 70 71 Examples: 72 >>> diff(parse_one("a + b"), parse_one("a + c")) 73 [ 74 Remove(expression=(COLUMN this: (IDENTIFIER this: b, quoted: False))), 75 Insert(expression=(COLUMN this: (IDENTIFIER this: c, quoted: False))), 76 Keep( 77 source=(ADD this: ...), 78 target=(ADD this: ...) 79 ), 80 Keep( 81 source=(COLUMN this: (IDENTIFIER this: a, quoted: False)), 82 target=(COLUMN this: (IDENTIFIER this: a, quoted: False)) 83 ), 84 ] 85 86 Args: 87 source: the source expression. 88 target: the target expression against which the diff should be calculated. 89 matchings: the list of pre-matched node pairs which is used to help the algorithm's 90 heuristics produce better results for subtrees that are known by a caller to be matching. 91 Note: expression references in this list must refer to the same node objects that are 92 referenced in source / target trees. 93 delta_only: excludes all `Keep` nodes from the diff. 94 kwargs: additional arguments to pass to the ChangeDistiller instance. 95 96 Returns: 97 the list of Insert, Remove, Move, Update and Keep objects for each node in the source and the 98 target expression trees. This list represents a sequence of steps needed to transform the source 99 expression tree into the target one. 100 """ 101 matchings = matchings or [] 102 matching_ids = {id(n) for pair in matchings for n in pair} 103 104 def compute_node_mappings( 105 original: exp.Expression, copy: exp.Expression 106 ) -> t.Dict[int, exp.Expression]: 107 return { 108 id(old_node): new_node 109 for old_node, new_node in zip(original.walk(), copy.walk()) 110 if id(old_node) in matching_ids 111 } 112 113 source_copy = source.copy() 114 target_copy = target.copy() 115 116 node_mappings = { 117 **compute_node_mappings(source, source_copy), 118 **compute_node_mappings(target, target_copy), 119 } 120 matchings_copy = [(node_mappings[id(s)], node_mappings[id(t)]) for s, t in matchings] 121 122 return ChangeDistiller(**kwargs).diff( 123 source_copy, 124 target_copy, 125 matchings=matchings_copy, 126 delta_only=delta_only, 127 ) 128 129 130# The expression types for which Update edits are allowed. 131UPDATABLE_EXPRESSION_TYPES = ( 132 exp.Alias, 133 exp.Boolean, 134 exp.Column, 135 exp.DataType, 136 exp.Lambda, 137 exp.Literal, 138 exp.Table, 139 exp.Window, 140) 141 142IGNORED_LEAF_EXPRESSION_TYPES = (exp.Identifier,) 143 144 145class ChangeDistiller: 146 """ 147 The implementation of the Change Distiller algorithm described by Beat Fluri and Martin Pinzger in 148 their paper https://ieeexplore.ieee.org/document/4339230, which in turn is based on the algorithm by 149 Chawathe et al. described in http://ilpubs.stanford.edu:8090/115/1/1995-46.pdf. 150 """ 151 152 def __init__(self, f: float = 0.6, t: float = 0.6) -> None: 153 self.f = f 154 self.t = t 155 self._sql_generator = Dialect().generator() 156 157 def diff( 158 self, 159 source: exp.Expression, 160 target: exp.Expression, 161 matchings: t.List[t.Tuple[exp.Expression, exp.Expression]] | None = None, 162 delta_only: bool = False, 163 ) -> t.List[Edit]: 164 matchings = matchings or [] 165 pre_matched_nodes = {id(s): id(t) for s, t in matchings} 166 if len({n for pair in pre_matched_nodes.items() for n in pair}) != 2 * len(matchings): 167 raise ValueError("Each node can be referenced at most once in the list of matchings") 168 169 self._source = source 170 self._target = target 171 self._source_index = { 172 id(n): n for n in self._source.bfs() if not isinstance(n, IGNORED_LEAF_EXPRESSION_TYPES) 173 } 174 self._target_index = { 175 id(n): n for n in self._target.bfs() if not isinstance(n, IGNORED_LEAF_EXPRESSION_TYPES) 176 } 177 self._unmatched_source_nodes = set(self._source_index) - set(pre_matched_nodes) 178 self._unmatched_target_nodes = set(self._target_index) - set(pre_matched_nodes.values()) 179 self._bigram_histo_cache: t.Dict[int, t.DefaultDict[str, int]] = {} 180 181 matching_set = self._compute_matching_set() | {(s, t) for s, t in pre_matched_nodes.items()} 182 return self._generate_edit_script(matching_set, delta_only) 183 184 def _generate_edit_script( 185 self, 186 matching_set: t.Set[t.Tuple[int, int]], 187 delta_only: bool, 188 ) -> t.List[Edit]: 189 edit_script: t.List[Edit] = [] 190 for removed_node_id in self._unmatched_source_nodes: 191 edit_script.append(Remove(self._source_index[removed_node_id])) 192 for inserted_node_id in self._unmatched_target_nodes: 193 edit_script.append(Insert(self._target_index[inserted_node_id])) 194 for kept_source_node_id, kept_target_node_id in matching_set: 195 source_node = self._source_index[kept_source_node_id] 196 target_node = self._target_index[kept_target_node_id] 197 if ( 198 not isinstance(source_node, UPDATABLE_EXPRESSION_TYPES) 199 or source_node == target_node 200 ): 201 edit_script.extend( 202 self._generate_move_edits(source_node, target_node, matching_set) 203 ) 204 if not delta_only: 205 edit_script.append(Keep(source_node, target_node)) 206 else: 207 edit_script.append(Update(source_node, target_node)) 208 209 return edit_script 210 211 def _generate_move_edits( 212 self, source: exp.Expression, target: exp.Expression, matching_set: t.Set[t.Tuple[int, int]] 213 ) -> t.List[Move]: 214 source_args = [id(e) for e in _expression_only_args(source)] 215 target_args = [id(e) for e in _expression_only_args(target)] 216 217 args_lcs = set(_lcs(source_args, target_args, lambda l, r: (l, r) in matching_set)) 218 219 move_edits = [] 220 for a in source_args: 221 if a not in args_lcs and a not in self._unmatched_source_nodes: 222 move_edits.append(Move(self._source_index[a])) 223 224 return move_edits 225 226 def _compute_matching_set(self) -> t.Set[t.Tuple[int, int]]: 227 leaves_matching_set = self._compute_leaf_matching_set() 228 matching_set = leaves_matching_set.copy() 229 230 ordered_unmatched_source_nodes = { 231 id(n): None for n in self._source.bfs() if id(n) in self._unmatched_source_nodes 232 } 233 ordered_unmatched_target_nodes = { 234 id(n): None for n in self._target.bfs() if id(n) in self._unmatched_target_nodes 235 } 236 237 for source_node_id in ordered_unmatched_source_nodes: 238 for target_node_id in ordered_unmatched_target_nodes: 239 source_node = self._source_index[source_node_id] 240 target_node = self._target_index[target_node_id] 241 if _is_same_type(source_node, target_node): 242 source_leaf_ids = {id(l) for l in _get_leaves(source_node)} 243 target_leaf_ids = {id(l) for l in _get_leaves(target_node)} 244 245 max_leaves_num = max(len(source_leaf_ids), len(target_leaf_ids)) 246 if max_leaves_num: 247 common_leaves_num = sum( 248 1 if s in source_leaf_ids and t in target_leaf_ids else 0 249 for s, t in leaves_matching_set 250 ) 251 leaf_similarity_score = common_leaves_num / max_leaves_num 252 else: 253 leaf_similarity_score = 0.0 254 255 adjusted_t = ( 256 self.t if min(len(source_leaf_ids), len(target_leaf_ids)) > 4 else 0.4 257 ) 258 259 if leaf_similarity_score >= 0.8 or ( 260 leaf_similarity_score >= adjusted_t 261 and self._dice_coefficient(source_node, target_node) >= self.f 262 ): 263 matching_set.add((source_node_id, target_node_id)) 264 self._unmatched_source_nodes.remove(source_node_id) 265 self._unmatched_target_nodes.remove(target_node_id) 266 ordered_unmatched_target_nodes.pop(target_node_id, None) 267 break 268 269 return matching_set 270 271 def _compute_leaf_matching_set(self) -> t.Set[t.Tuple[int, int]]: 272 candidate_matchings: t.List[t.Tuple[float, int, int, exp.Expression, exp.Expression]] = [] 273 source_leaves = list(_get_leaves(self._source)) 274 target_leaves = list(_get_leaves(self._target)) 275 for source_leaf in source_leaves: 276 for target_leaf in target_leaves: 277 if _is_same_type(source_leaf, target_leaf): 278 similarity_score = self._dice_coefficient(source_leaf, target_leaf) 279 if similarity_score >= self.f: 280 heappush( 281 candidate_matchings, 282 ( 283 -similarity_score, 284 -_parent_similarity_score(source_leaf, target_leaf), 285 len(candidate_matchings), 286 source_leaf, 287 target_leaf, 288 ), 289 ) 290 291 # Pick best matchings based on the highest score 292 matching_set = set() 293 while candidate_matchings: 294 _, _, _, source_leaf, target_leaf = heappop(candidate_matchings) 295 if ( 296 id(source_leaf) in self._unmatched_source_nodes 297 and id(target_leaf) in self._unmatched_target_nodes 298 ): 299 matching_set.add((id(source_leaf), id(target_leaf))) 300 self._unmatched_source_nodes.remove(id(source_leaf)) 301 self._unmatched_target_nodes.remove(id(target_leaf)) 302 303 return matching_set 304 305 def _dice_coefficient(self, source: exp.Expression, target: exp.Expression) -> float: 306 source_histo = self._bigram_histo(source) 307 target_histo = self._bigram_histo(target) 308 309 total_grams = sum(source_histo.values()) + sum(target_histo.values()) 310 if not total_grams: 311 return 1.0 if source == target else 0.0 312 313 overlap_len = 0 314 overlapping_grams = set(source_histo) & set(target_histo) 315 for g in overlapping_grams: 316 overlap_len += min(source_histo[g], target_histo[g]) 317 318 return 2 * overlap_len / total_grams 319 320 def _bigram_histo(self, expression: exp.Expression) -> t.DefaultDict[str, int]: 321 if id(expression) in self._bigram_histo_cache: 322 return self._bigram_histo_cache[id(expression)] 323 324 expression_str = self._sql_generator.generate(expression) 325 count = max(0, len(expression_str) - 1) 326 bigram_histo: t.DefaultDict[str, int] = defaultdict(int) 327 for i in range(count): 328 bigram_histo[expression_str[i : i + 2]] += 1 329 330 self._bigram_histo_cache[id(expression)] = bigram_histo 331 return bigram_histo 332 333 334def _get_leaves(expression: exp.Expression) -> t.Iterator[exp.Expression]: 335 has_child_exprs = False 336 337 for node in expression.iter_expressions(): 338 if not isinstance(node, IGNORED_LEAF_EXPRESSION_TYPES): 339 has_child_exprs = True 340 yield from _get_leaves(node) 341 342 if not has_child_exprs: 343 yield expression 344 345 346def _is_same_type(source: exp.Expression, target: exp.Expression) -> bool: 347 if type(source) is type(target): 348 if isinstance(source, exp.Join): 349 return source.args.get("side") == target.args.get("side") 350 351 if isinstance(source, exp.Anonymous): 352 return source.this == target.this 353 354 return True 355 356 return False 357 358 359def _parent_similarity_score( 360 source: t.Optional[exp.Expression], target: t.Optional[exp.Expression] 361) -> int: 362 if source is None or target is None or type(source) is not type(target): 363 return 0 364 365 return 1 + _parent_similarity_score(source.parent, target.parent) 366 367 368def _expression_only_args(expression: exp.Expression) -> t.List[exp.Expression]: 369 args: t.List[t.Union[exp.Expression, t.List]] = [] 370 if expression: 371 for a in expression.args.values(): 372 args.extend(ensure_list(a)) 373 return [ 374 a 375 for a in args 376 if isinstance(a, exp.Expression) and not isinstance(a, IGNORED_LEAF_EXPRESSION_TYPES) 377 ] 378 379 380def _lcs( 381 seq_a: t.Sequence[T], seq_b: t.Sequence[T], equal: t.Callable[[T, T], bool] 382) -> t.Sequence[t.Optional[T]]: 383 """Calculates the longest common subsequence""" 384 385 len_a = len(seq_a) 386 len_b = len(seq_b) 387 lcs_result = [[None] * (len_b + 1) for i in range(len_a + 1)] 388 389 for i in range(len_a + 1): 390 for j in range(len_b + 1): 391 if i == 0 or j == 0: 392 lcs_result[i][j] = [] # type: ignore 393 elif equal(seq_a[i - 1], seq_b[j - 1]): 394 lcs_result[i][j] = lcs_result[i - 1][j - 1] + [seq_a[i - 1]] # type: ignore 395 else: 396 lcs_result[i][j] = ( 397 lcs_result[i - 1][j] 398 if len(lcs_result[i - 1][j]) > len(lcs_result[i][j - 1]) # type: ignore 399 else lcs_result[i][j - 1] 400 ) 401 402 return lcs_result[len_a][len_b] # type: ignore
19@dataclass(frozen=True) 20class Insert: 21 """Indicates that a new node has been inserted""" 22 23 expression: exp.Expression
Indicates that a new node has been inserted
26@dataclass(frozen=True) 27class Remove: 28 """Indicates that an existing node has been removed""" 29 30 expression: exp.Expression
Indicates that an existing node has been removed
33@dataclass(frozen=True) 34class Move: 35 """Indicates that an existing node's position within the tree has changed""" 36 37 expression: exp.Expression
Indicates that an existing node's position within the tree has changed
40@dataclass(frozen=True) 41class Update: 42 """Indicates that an existing node has been updated""" 43 44 source: exp.Expression 45 target: exp.Expression
Indicates that an existing node has been updated
48@dataclass(frozen=True) 49class Keep: 50 """Indicates that an existing node hasn't been changed""" 51 52 source: exp.Expression 53 target: exp.Expression
Indicates that an existing node hasn't been changed
62def diff( 63 source: exp.Expression, 64 target: exp.Expression, 65 matchings: t.List[t.Tuple[exp.Expression, exp.Expression]] | None = None, 66 delta_only: bool = False, 67 **kwargs: t.Any, 68) -> t.List[Edit]: 69 """ 70 Returns the list of changes between the source and the target expressions. 71 72 Examples: 73 >>> diff(parse_one("a + b"), parse_one("a + c")) 74 [ 75 Remove(expression=(COLUMN this: (IDENTIFIER this: b, quoted: False))), 76 Insert(expression=(COLUMN this: (IDENTIFIER this: c, quoted: False))), 77 Keep( 78 source=(ADD this: ...), 79 target=(ADD this: ...) 80 ), 81 Keep( 82 source=(COLUMN this: (IDENTIFIER this: a, quoted: False)), 83 target=(COLUMN this: (IDENTIFIER this: a, quoted: False)) 84 ), 85 ] 86 87 Args: 88 source: the source expression. 89 target: the target expression against which the diff should be calculated. 90 matchings: the list of pre-matched node pairs which is used to help the algorithm's 91 heuristics produce better results for subtrees that are known by a caller to be matching. 92 Note: expression references in this list must refer to the same node objects that are 93 referenced in source / target trees. 94 delta_only: excludes all `Keep` nodes from the diff. 95 kwargs: additional arguments to pass to the ChangeDistiller instance. 96 97 Returns: 98 the list of Insert, Remove, Move, Update and Keep objects for each node in the source and the 99 target expression trees. This list represents a sequence of steps needed to transform the source 100 expression tree into the target one. 101 """ 102 matchings = matchings or [] 103 matching_ids = {id(n) for pair in matchings for n in pair} 104 105 def compute_node_mappings( 106 original: exp.Expression, copy: exp.Expression 107 ) -> t.Dict[int, exp.Expression]: 108 return { 109 id(old_node): new_node 110 for old_node, new_node in zip(original.walk(), copy.walk()) 111 if id(old_node) in matching_ids 112 } 113 114 source_copy = source.copy() 115 target_copy = target.copy() 116 117 node_mappings = { 118 **compute_node_mappings(source, source_copy), 119 **compute_node_mappings(target, target_copy), 120 } 121 matchings_copy = [(node_mappings[id(s)], node_mappings[id(t)]) for s, t in matchings] 122 123 return ChangeDistiller(**kwargs).diff( 124 source_copy, 125 target_copy, 126 matchings=matchings_copy, 127 delta_only=delta_only, 128 )
Returns the list of changes between the source and the target expressions.
Examples:
>>> diff(parse_one("a + b"), parse_one("a + c")) [ Remove(expression=(COLUMN this: (IDENTIFIER this: b, quoted: False))), Insert(expression=(COLUMN this: (IDENTIFIER this: c, quoted: False))), Keep( source=(ADD this: ...), target=(ADD this: ...) ), Keep( source=(COLUMN this: (IDENTIFIER this: a, quoted: False)), target=(COLUMN this: (IDENTIFIER this: a, quoted: False)) ), ]
Arguments:
- source: the source expression.
- target: the target expression against which the diff should be calculated.
- matchings: the list of pre-matched node pairs which is used to help the algorithm's heuristics produce better results for subtrees that are known by a caller to be matching. Note: expression references in this list must refer to the same node objects that are referenced in source / target trees.
- delta_only: excludes all
Keep
nodes from the diff. - kwargs: additional arguments to pass to the ChangeDistiller instance.
Returns:
the list of Insert, Remove, Move, Update and Keep objects for each node in the source and the target expression trees. This list represents a sequence of steps needed to transform the source expression tree into the target one.
146class ChangeDistiller: 147 """ 148 The implementation of the Change Distiller algorithm described by Beat Fluri and Martin Pinzger in 149 their paper https://ieeexplore.ieee.org/document/4339230, which in turn is based on the algorithm by 150 Chawathe et al. described in http://ilpubs.stanford.edu:8090/115/1/1995-46.pdf. 151 """ 152 153 def __init__(self, f: float = 0.6, t: float = 0.6) -> None: 154 self.f = f 155 self.t = t 156 self._sql_generator = Dialect().generator() 157 158 def diff( 159 self, 160 source: exp.Expression, 161 target: exp.Expression, 162 matchings: t.List[t.Tuple[exp.Expression, exp.Expression]] | None = None, 163 delta_only: bool = False, 164 ) -> t.List[Edit]: 165 matchings = matchings or [] 166 pre_matched_nodes = {id(s): id(t) for s, t in matchings} 167 if len({n for pair in pre_matched_nodes.items() for n in pair}) != 2 * len(matchings): 168 raise ValueError("Each node can be referenced at most once in the list of matchings") 169 170 self._source = source 171 self._target = target 172 self._source_index = { 173 id(n): n for n in self._source.bfs() if not isinstance(n, IGNORED_LEAF_EXPRESSION_TYPES) 174 } 175 self._target_index = { 176 id(n): n for n in self._target.bfs() if not isinstance(n, IGNORED_LEAF_EXPRESSION_TYPES) 177 } 178 self._unmatched_source_nodes = set(self._source_index) - set(pre_matched_nodes) 179 self._unmatched_target_nodes = set(self._target_index) - set(pre_matched_nodes.values()) 180 self._bigram_histo_cache: t.Dict[int, t.DefaultDict[str, int]] = {} 181 182 matching_set = self._compute_matching_set() | {(s, t) for s, t in pre_matched_nodes.items()} 183 return self._generate_edit_script(matching_set, delta_only) 184 185 def _generate_edit_script( 186 self, 187 matching_set: t.Set[t.Tuple[int, int]], 188 delta_only: bool, 189 ) -> t.List[Edit]: 190 edit_script: t.List[Edit] = [] 191 for removed_node_id in self._unmatched_source_nodes: 192 edit_script.append(Remove(self._source_index[removed_node_id])) 193 for inserted_node_id in self._unmatched_target_nodes: 194 edit_script.append(Insert(self._target_index[inserted_node_id])) 195 for kept_source_node_id, kept_target_node_id in matching_set: 196 source_node = self._source_index[kept_source_node_id] 197 target_node = self._target_index[kept_target_node_id] 198 if ( 199 not isinstance(source_node, UPDATABLE_EXPRESSION_TYPES) 200 or source_node == target_node 201 ): 202 edit_script.extend( 203 self._generate_move_edits(source_node, target_node, matching_set) 204 ) 205 if not delta_only: 206 edit_script.append(Keep(source_node, target_node)) 207 else: 208 edit_script.append(Update(source_node, target_node)) 209 210 return edit_script 211 212 def _generate_move_edits( 213 self, source: exp.Expression, target: exp.Expression, matching_set: t.Set[t.Tuple[int, int]] 214 ) -> t.List[Move]: 215 source_args = [id(e) for e in _expression_only_args(source)] 216 target_args = [id(e) for e in _expression_only_args(target)] 217 218 args_lcs = set(_lcs(source_args, target_args, lambda l, r: (l, r) in matching_set)) 219 220 move_edits = [] 221 for a in source_args: 222 if a not in args_lcs and a not in self._unmatched_source_nodes: 223 move_edits.append(Move(self._source_index[a])) 224 225 return move_edits 226 227 def _compute_matching_set(self) -> t.Set[t.Tuple[int, int]]: 228 leaves_matching_set = self._compute_leaf_matching_set() 229 matching_set = leaves_matching_set.copy() 230 231 ordered_unmatched_source_nodes = { 232 id(n): None for n in self._source.bfs() if id(n) in self._unmatched_source_nodes 233 } 234 ordered_unmatched_target_nodes = { 235 id(n): None for n in self._target.bfs() if id(n) in self._unmatched_target_nodes 236 } 237 238 for source_node_id in ordered_unmatched_source_nodes: 239 for target_node_id in ordered_unmatched_target_nodes: 240 source_node = self._source_index[source_node_id] 241 target_node = self._target_index[target_node_id] 242 if _is_same_type(source_node, target_node): 243 source_leaf_ids = {id(l) for l in _get_leaves(source_node)} 244 target_leaf_ids = {id(l) for l in _get_leaves(target_node)} 245 246 max_leaves_num = max(len(source_leaf_ids), len(target_leaf_ids)) 247 if max_leaves_num: 248 common_leaves_num = sum( 249 1 if s in source_leaf_ids and t in target_leaf_ids else 0 250 for s, t in leaves_matching_set 251 ) 252 leaf_similarity_score = common_leaves_num / max_leaves_num 253 else: 254 leaf_similarity_score = 0.0 255 256 adjusted_t = ( 257 self.t if min(len(source_leaf_ids), len(target_leaf_ids)) > 4 else 0.4 258 ) 259 260 if leaf_similarity_score >= 0.8 or ( 261 leaf_similarity_score >= adjusted_t 262 and self._dice_coefficient(source_node, target_node) >= self.f 263 ): 264 matching_set.add((source_node_id, target_node_id)) 265 self._unmatched_source_nodes.remove(source_node_id) 266 self._unmatched_target_nodes.remove(target_node_id) 267 ordered_unmatched_target_nodes.pop(target_node_id, None) 268 break 269 270 return matching_set 271 272 def _compute_leaf_matching_set(self) -> t.Set[t.Tuple[int, int]]: 273 candidate_matchings: t.List[t.Tuple[float, int, int, exp.Expression, exp.Expression]] = [] 274 source_leaves = list(_get_leaves(self._source)) 275 target_leaves = list(_get_leaves(self._target)) 276 for source_leaf in source_leaves: 277 for target_leaf in target_leaves: 278 if _is_same_type(source_leaf, target_leaf): 279 similarity_score = self._dice_coefficient(source_leaf, target_leaf) 280 if similarity_score >= self.f: 281 heappush( 282 candidate_matchings, 283 ( 284 -similarity_score, 285 -_parent_similarity_score(source_leaf, target_leaf), 286 len(candidate_matchings), 287 source_leaf, 288 target_leaf, 289 ), 290 ) 291 292 # Pick best matchings based on the highest score 293 matching_set = set() 294 while candidate_matchings: 295 _, _, _, source_leaf, target_leaf = heappop(candidate_matchings) 296 if ( 297 id(source_leaf) in self._unmatched_source_nodes 298 and id(target_leaf) in self._unmatched_target_nodes 299 ): 300 matching_set.add((id(source_leaf), id(target_leaf))) 301 self._unmatched_source_nodes.remove(id(source_leaf)) 302 self._unmatched_target_nodes.remove(id(target_leaf)) 303 304 return matching_set 305 306 def _dice_coefficient(self, source: exp.Expression, target: exp.Expression) -> float: 307 source_histo = self._bigram_histo(source) 308 target_histo = self._bigram_histo(target) 309 310 total_grams = sum(source_histo.values()) + sum(target_histo.values()) 311 if not total_grams: 312 return 1.0 if source == target else 0.0 313 314 overlap_len = 0 315 overlapping_grams = set(source_histo) & set(target_histo) 316 for g in overlapping_grams: 317 overlap_len += min(source_histo[g], target_histo[g]) 318 319 return 2 * overlap_len / total_grams 320 321 def _bigram_histo(self, expression: exp.Expression) -> t.DefaultDict[str, int]: 322 if id(expression) in self._bigram_histo_cache: 323 return self._bigram_histo_cache[id(expression)] 324 325 expression_str = self._sql_generator.generate(expression) 326 count = max(0, len(expression_str) - 1) 327 bigram_histo: t.DefaultDict[str, int] = defaultdict(int) 328 for i in range(count): 329 bigram_histo[expression_str[i : i + 2]] += 1 330 331 self._bigram_histo_cache[id(expression)] = bigram_histo 332 return bigram_histo
The implementation of the Change Distiller algorithm described by Beat Fluri and Martin Pinzger in their paper https://ieeexplore.ieee.org/document/4339230, which in turn is based on the algorithm by Chawathe et al. described in http://ilpubs.stanford.edu:8090/115/1/1995-46.pdf.
158 def diff( 159 self, 160 source: exp.Expression, 161 target: exp.Expression, 162 matchings: t.List[t.Tuple[exp.Expression, exp.Expression]] | None = None, 163 delta_only: bool = False, 164 ) -> t.List[Edit]: 165 matchings = matchings or [] 166 pre_matched_nodes = {id(s): id(t) for s, t in matchings} 167 if len({n for pair in pre_matched_nodes.items() for n in pair}) != 2 * len(matchings): 168 raise ValueError("Each node can be referenced at most once in the list of matchings") 169 170 self._source = source 171 self._target = target 172 self._source_index = { 173 id(n): n for n in self._source.bfs() if not isinstance(n, IGNORED_LEAF_EXPRESSION_TYPES) 174 } 175 self._target_index = { 176 id(n): n for n in self._target.bfs() if not isinstance(n, IGNORED_LEAF_EXPRESSION_TYPES) 177 } 178 self._unmatched_source_nodes = set(self._source_index) - set(pre_matched_nodes) 179 self._unmatched_target_nodes = set(self._target_index) - set(pre_matched_nodes.values()) 180 self._bigram_histo_cache: t.Dict[int, t.DefaultDict[str, int]] = {} 181 182 matching_set = self._compute_matching_set() | {(s, t) for s, t in pre_matched_nodes.items()} 183 return self._generate_edit_script(matching_set, delta_only)