Reformat Mononoke thrift files to match arc f

Summary: I have my editor set up to format on save - let's unify this on the standard FB format for thrift files, so that I don't create junk.

Reviewed By: ahornby

Differential Revision: D30285082

fbshipit-source-id: 17b09635a2473174a92e29bb042432dbac44865a

eden/mononoke/blobstore/if/memcache_lock.thrift

@@ -6,6 +6,6 @@
  */
 
 union LockState {
-  1: string locked_by,
+  1: string locked_by;
-  2: string uploaded_key,
+  2: string uploaded_key;
 }

eden/mononoke/bonsai_globalrev_mapping/if/bonsai_globalrev_mapping.thrift

@@ -9,11 +9,11 @@ include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 
 # Memcache constants. Should be change when we want to invalidate memcache
 # entries
-const i32 MC_CODEVER = 0
+const i32 MC_CODEVER = 0;
-const i32 MC_SITEVER = 0
+const i32 MC_SITEVER = 0;
 
 struct BonsaiGlobalrevMappingEntry {
-  1: required i32 repo_id,
+  1: required i32 repo_id;
-  2: required mononoke_types_thrift.ChangesetId bcs_id,
+  2: required mononoke_types_thrift.ChangesetId bcs_id;
-  3: required i64 globalrev,
+  3: required i64 globalrev;
 }

eden/mononoke/bonsai_hg_mapping/if/bonsai_hg_mapping.thrift

@@ -10,13 +10,13 @@ include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 
 # Memcache constants. Should be change when we want to invalidate memcache
 # entries
-const i32 MC_CODEVER = 0
+const i32 MC_CODEVER = 0;
-const i32 MC_SITEVER = 3
+const i32 MC_SITEVER = 3;
 
 typedef i32 RepoId (rust.newtype)
 
 struct BonsaiHgMappingEntry {
-  1: required RepoId repo_id,
+  1: required RepoId repo_id;
-  2: required mononoke_types_thrift.ChangesetId bcs_id,
+  2: required mononoke_types_thrift.ChangesetId bcs_id;
-  3: required mercurial_thrift.HgNodeHash hg_cs_id,
+  3: required mercurial_thrift.HgNodeHash hg_cs_id;
 }

eden/mononoke/bonsai_svnrev_mapping/if/bonsai_svnrev_mapping.thrift

@@ -9,11 +9,11 @@ include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 
 # Memcache constants. Should be change when we want to invalidate memcache
 # entries
-const i32 MC_CODEVER = 0
+const i32 MC_CODEVER = 0;
-const i32 MC_SITEVER = 0
+const i32 MC_SITEVER = 0;
 
 struct BonsaiSvnrevMappingEntry {
-  1: required i32 repo_id,
+  1: required i32 repo_id;
-  2: required mononoke_types_thrift.ChangesetId bcs_id,
+  2: required mononoke_types_thrift.ChangesetId bcs_id;
-  3: required i64 svnrev,
+  3: required i64 svnrev;
 }

eden/mononoke/changesets/if/changeset_entry.thrift

@@ -9,17 +9,17 @@ include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 
 # Memcache constants. Should be change when we want to invalidate memcache
 # entries
-const i32 MC_CODEVER = 0
+const i32 MC_CODEVER = 0;
-const i32 MC_SITEVER = 0
+const i32 MC_SITEVER = 0;
 
 typedef i32 RepoId (rust.newtype)
 
- // Thrift does not support unsigned, so using i64 here
+// Thrift does not support unsigned, so using i64 here
 typedef i64 GenerationNum (rust.newtype)
 
 struct ChangesetEntry {
-  1: required RepoId repo_id,
+  1: required RepoId repo_id;
-  2: required mononoke_types_thrift.ChangesetId cs_id,
+  2: required mononoke_types_thrift.ChangesetId cs_id;
-  3: required list<mononoke_types_thrift.ChangesetId> parents,
+  3: required list<mononoke_types_thrift.ChangesetId> parents;
-  4: required GenerationNum gen,
+  4: required GenerationNum gen;
 }

eden/mononoke/derived_data/changeset_info/if/changeset_info_thrift.thrift

@@ -16,14 +16,16 @@ include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 // reason the commit is being fetched.
 struct ChangesetInfo {
   // Changeset id of the source Bonsai changeset
-  1: mononoke_types_thrift.ChangesetId changeset_id,
+  1: mononoke_types_thrift.ChangesetId changeset_id;
-  2: list<mononoke_types_thrift.ChangesetId> parents,
+  2: list<mononoke_types_thrift.ChangesetId> parents;
-  3: string author,
+  3: string author;
-  4: mononoke_types_thrift.DateTime author_date,
+  4: mononoke_types_thrift.DateTime author_date;
-  5: optional string committer,
+  5: optional string committer;
-  6: optional mononoke_types_thrift.DateTime committer_date,
+  6: optional mononoke_types_thrift.DateTime committer_date;
-  7: ChangesetMessage message,
+  7: ChangesetMessage message;
-  8: map<string, binary> (rust.type = "sorted_vector_map::SortedVectorMap") extra,
+  8: map<string, binary> (
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) extra;
 }
 
 // Commit message is represented by a separate union of formats for the future
@@ -31,5 +33,5 @@ struct ChangesetInfo {
 // At some point we may like to store large commit messages as separate blobs to
 // make fetching changesets faster if there is no need in the whole description.
 union ChangesetMessage {
-  1: string message,
+  1: string message;
 }

eden/mononoke/filenodes/if/filenodes.thrift

@@ -10,26 +10,26 @@ include "eden/mononoke/mercurial/types/if/mercurial_thrift.thrift"
 
 # Memcache constants. Should be change when we want to invalidate memcache
 # entries
-const i32 MC_CODEVER = 3
+const i32 MC_CODEVER = 3;
-const i32 MC_SITEVER = 1
+const i32 MC_SITEVER = 1;
 
 union FilenodeInfoList {
-  1: list<FilenodeInfo> Data,
+  1: list<FilenodeInfo> Data;
-  2: list<i64> Pointers,
+  2: list<i64> Pointers;
   // This actual value is ignored
-  3: byte TooBig,
+  3: byte TooBig;
 }
 
 struct FilenodeInfo {
   // 1 was used previously.
-  2: required mercurial_thrift.HgNodeHash filenode,
+  2: required mercurial_thrift.HgNodeHash filenode;
-  3: optional mercurial_thrift.HgNodeHash p1,
+  3: optional mercurial_thrift.HgNodeHash p1;
-  4: optional mercurial_thrift.HgNodeHash p2,
+  4: optional mercurial_thrift.HgNodeHash p2;
-  5: optional FilenodeCopyFrom copyfrom,
+  5: optional FilenodeCopyFrom copyfrom;
-  6: required mercurial_thrift.HgNodeHash linknode,
+  6: required mercurial_thrift.HgNodeHash linknode;
 }
 
 struct FilenodeCopyFrom {
-  1: required mononoke_types_thrift.RepoPath path,
+  1: required mononoke_types_thrift.RepoPath path;
-  2: required mercurial_thrift.HgNodeHash filenode,
+  2: required mercurial_thrift.HgNodeHash filenode;
 }

eden/mononoke/git/git_types/if/git_types_thrift.thrift

@@ -8,22 +8,22 @@
 include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 
 struct BlobHandle {
-  1: mononoke_types_thrift.GitSha1 oid,
+  1: mononoke_types_thrift.GitSha1 oid;
-  2: i64 size,
+  2: i64 size;
-  3: mononoke_types_thrift.FileType file_type,
+  3: mononoke_types_thrift.FileType file_type;
 }
 
 struct TreeHandle {
-  1: mononoke_types_thrift.GitSha1 oid,
+  1: mononoke_types_thrift.GitSha1 oid;
-  2: i64 size,
+  2: i64 size;
 }
 
 union TreeMember {
-  1: BlobHandle Blob,
+  1: BlobHandle Blob;
-  2: TreeHandle Tree,
+  2: TreeHandle Tree;
 }
 
 struct Tree {
-  1: TreeHandle handle,
+  1: TreeHandle handle;
-  2: map<mononoke_types_thrift.MPathElement, TreeMember> members,
+  2: map<mononoke_types_thrift.MPathElement, TreeMember> members;
 }

eden/mononoke/mercurial/types/if/mercurial_thrift.thrift

@@ -12,36 +12,36 @@ typedef mononoke_types_thrift.Sha1 HgNodeHash (rust.newtype)
 // Changeset contents are stored inline.
 struct HgChangesetEnvelope {
   // The node ID is expected to match the contents exactly.
-  1: required HgNodeHash node_id,
+  1: required HgNodeHash node_id;
-  2: optional HgNodeHash p1,
+  2: optional HgNodeHash p1;
-  3: optional HgNodeHash p2,
+  3: optional HgNodeHash p2;
   // These contents are exactly as they would be serialized by Mercurial.
-  4: optional binary contents,
+  4: optional binary contents;
 }
 
 // Manifest contents are expected to generally be small, so they're stored
 // inline in the envelope. There's also no real dedup possible between native
 // Mononoke data structures and these ones.
 struct HgManifestEnvelope {
-  1: required HgNodeHash node_id,
+  1: required HgNodeHash node_id;
-  2: optional HgNodeHash p1,
+  2: optional HgNodeHash p1;
-  3: optional HgNodeHash p2,
+  3: optional HgNodeHash p2;
   // Root tree manifest nodes can have node IDs that don't match the contents.
   // That is required for lookups, but it means that in the event of recovery
   // from a disaster, hash consistency can't be checked. The computed node ID
   // is stored to allow that to happen.
-  4: required HgNodeHash computed_node_id,
+  4: required HgNodeHash computed_node_id;
   // These contents are exactly as they would be serialized by Mercurial.
-  5: optional binary contents,
+  5: optional binary contents;
 }
 
 struct HgFileEnvelope {
-  1: required HgNodeHash node_id,
+  1: required HgNodeHash node_id;
-  2: optional HgNodeHash p1,
+  2: optional HgNodeHash p1;
-  3: optional HgNodeHash p2,
+  3: optional HgNodeHash p2;
-  4: optional mononoke_types_thrift.ContentId content_id,
+  4: optional mononoke_types_thrift.ContentId content_id;
   // content_size is a u64 stored as an i64, and doesn't include the size of
   // the metadata
-  5: required i64 content_size,
+  5: required i64 content_size;
-  6: optional binary metadata,
+  6: optional binary metadata;
 }

eden/mononoke/microwave/if/microwave.thrift

@@ -9,31 +9,31 @@ include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 include "eden/mononoke/mercurial/types/if/mercurial_thrift.thrift"
 
 // Code version constant -- update to invalidate saved state.
-const i32 CODEVER = 1
+const i32 CODEVER = 1;
 
 struct FilenodeSnapshot {
   // Note: required fields are enforced at runtime here (to prevent Thift from
   // giving us garbage values and calling those acceptable).
-  1: optional mononoke_types_thrift.RepoPath path,
+  1: optional mononoke_types_thrift.RepoPath path;
-  2: optional mercurial_thrift.HgNodeHash filenode,
+  2: optional mercurial_thrift.HgNodeHash filenode;
-  3: optional mercurial_thrift.HgNodeHash p1,
+  3: optional mercurial_thrift.HgNodeHash p1;
-  4: optional mercurial_thrift.HgNodeHash p2,
+  4: optional mercurial_thrift.HgNodeHash p2;
-  5: optional CopyInfoSnapshot copyfrom,
+  5: optional CopyInfoSnapshot copyfrom;
-  6: optional mercurial_thrift.HgNodeHash linknode,
+  6: optional mercurial_thrift.HgNodeHash linknode;
 }
 
 struct CopyInfoSnapshot {
-  1: optional mononoke_types_thrift.RepoPath path,
+  1: optional mononoke_types_thrift.RepoPath path;
-  2: optional mercurial_thrift.HgNodeHash filenode,
+  2: optional mercurial_thrift.HgNodeHash filenode;
 }
 
 struct ChangesetSnapshot {
-  1: optional mononoke_types_thrift.ChangesetId cs_id,
+  1: optional mononoke_types_thrift.ChangesetId cs_id;
-  2: optional list<mononoke_types_thrift.ChangesetId> parents,
+  2: optional list<mononoke_types_thrift.ChangesetId> parents;
-  3: optional i64 gen,
+  3: optional i64 gen;
 }
 
 struct RepoSnapshot {
-  1: optional list<FilenodeSnapshot> filenodes,
+  1: optional list<FilenodeSnapshot> filenodes;
-  2: optional list<ChangesetSnapshot> changesets,
+  2: optional list<ChangesetSnapshot> changesets;
 }

eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift

@@ -24,7 +24,7 @@ typedef binary (rust.type = "bytes::Bytes") binary_bytes
 
 // Allow the hash type to change in the future.
 union IdType {
-  1: Blake2 Blake2,
+  1: Blake2 Blake2;
 } (rust.ord)
 
 typedef IdType ChangesetId (rust.newtype)
@@ -33,9 +33,9 @@ typedef IdType ContentChunkId (rust.newtype)
 typedef IdType RawBundle2Id (rust.newtype)
 typedef IdType FileUnodeId (rust.newtype)
 typedef IdType ManifestUnodeId (rust.newtype)
-typedef IdType DeletedManifestId(rust.newtype)
+typedef IdType DeletedManifestId (rust.newtype)
 typedef IdType FsnodeId (rust.newtype)
-typedef IdType SkeletonManifestId(rust.newtype)
+typedef IdType SkeletonManifestId (rust.newtype)
 typedef IdType MPathHash (rust.newtype)
 
 typedef IdType ContentMetadataId (rust.newtype)
@@ -51,19 +51,25 @@ typedef binary Sha1 (rust.newtype, rust.type = "smallvec::SmallVec<[u8; 20]>")
 
 // Other content alias types
 typedef binary Sha256 (rust.newtype, rust.type = "smallvec::SmallVec<[u8; 32]>")
-typedef binary GitSha1 (rust.newtype, rust.type = "smallvec::SmallVec<[u8; 20]>")
+typedef binary GitSha1 (
+  rust.newtype,
+  rust.type = "smallvec::SmallVec<[u8; 20]>",
+)
 
 // A path in a repo is stored as a list of elements. This is so that the sort
 // order of paths is the same as that of a tree traversal, so that deltas on
 // manifests can be applied in a streaming way.
-typedef binary MPathElement (rust.newtype, rust.type = "smallvec::SmallVec<[u8; 24]>")
+typedef binary MPathElement (
+  rust.newtype,
+  rust.type = "smallvec::SmallVec<[u8; 24]>",
+)
 typedef list<MPathElement> MPath (rust.newtype)
 
 union RepoPath {
   # Thrift language doesn't support void here, so put a dummy bool
-  1: bool RootPath,
+  1: bool RootPath;
-  2: MPath DirectoryPath,
+  2: MPath DirectoryPath;
-  3: MPath FilePath,
+  3: MPath FilePath;
 }
 
 // Parent ordering
@@ -101,29 +107,33 @@ union RepoPath {
 //   * Corollary: The file list in Mercurial is not pcf, so the Bonsai diff is
 //     computed separately.
 struct BonsaiChangeset {
-  1: required list<ChangesetId> parents,
+  1: required list<ChangesetId> parents;
-  2: string author,
+  2: string author;
-  3: optional DateTime author_date,
+  3: optional DateTime author_date;
   // Mercurial won't necessarily have a committer, so this is optional.
-  4: optional string committer,
+  4: optional string committer;
-  5: optional DateTime committer_date,
+  5: optional DateTime committer_date;
-  6: string message,
+  6: string message;
-  7: map<string, binary> (rust.type = "sorted_vector_map::SortedVectorMap") extra,
+  7: map<string, binary> (
-  8: map<MPath, FileChangeOpt> (rust.type = "sorted_vector_map::SortedVectorMap") file_changes,
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) extra;
+  8: map<MPath, FileChangeOpt> (
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) file_changes;
 }
 
 // DateTime fields do not have a reasonable default value! They must
 // always be required or optional.
 struct DateTime {
-  1: required i64 timestamp_secs,
+  1: required i64 timestamp_secs;
   // Timezones can go up to UTC+13 (which would be represented as -46800), so
   // an i16 can't fit them.
-  2: required i32 tz_offset_secs,
+  2: required i32 tz_offset_secs;
 }
 
 struct ContentChunkPointer {
-  1: ContentChunkId chunk_id,
+  1: ContentChunkId chunk_id;
-  2: i64 size,
+  2: i64 size;
 }
 
 // When a file is chunked, we reprsent it as a list of its chunks, as well as
@@ -133,24 +143,24 @@ struct ChunkedFileContents {
   // FileContents reprseentation in Mononoke, which would normally require
   // hashing the contents (but we obviously can't do that here, since we don't
   // have the contents).
-  1: ContentId content_id,
+  1: ContentId content_id;
-  2: list<ContentChunkPointer> chunks,
+  2: list<ContentChunkPointer> chunks;
 }
 
 union FileContents {
   // Plain uncompressed bytes - WYSIWYG.
-  1: binary_bytes Bytes,
+  1: binary_bytes Bytes;
   // References to Chunks (stored as FileContents, too).
-  2: ChunkedFileContents Chunked,
+  2: ChunkedFileContents Chunked;
 }
 
 union ContentChunk {
-  1: binary_bytes Bytes,
+  1: binary_bytes Bytes;
 }
 
 // Payload of object which is an alias
 union ContentAlias {
-  1: ContentId ContentId, // File content alias
+  1: ContentId ContentId; // File content alias
 }
 
 // Metadata about a file. This includes hahs aliases, or the file's size.
@@ -161,17 +171,17 @@ union ContentAlias {
 // value).
 struct ContentMetadata {
   // total_size is needed to make GitSha1 meaningful, but generally useful
-  1: optional i64 total_size,
+  1: optional i64 total_size;
   // ContentId we're providing metadata for
-  2: optional ContentId content_id,
+  2: optional ContentId content_id;
-  3: optional Sha1 sha1,
+  3: optional Sha1 sha1;
-  4: optional Sha256 sha256,
+  4: optional Sha256 sha256;
   // always object type "blob"
-  5: optional GitSha1 git_sha1,
+  5: optional GitSha1 git_sha1;
 }
 
 union RawBundle2 {
-  1: binary Bytes,
+  1: binary Bytes;
 }
 
 enum FileType {
@@ -185,90 +195,94 @@ struct FileChangeOpt {
   // At most one value can be present.
 
   // Changes to a tracked file
-  1: optional FileChange change,
+  1: optional FileChange change;
   // This is a change to an untracked file in a snapshot commit.
-  2: optional UntrackedFileChange untracked_change,
+  2: optional UntrackedFileChange untracked_change;
   // Present if this is a missing file in a snapshot commit.
-  3: optional UntrackedDeletion untracked_deletion,
+  3: optional UntrackedDeletion untracked_deletion;
 }
 
 struct UntrackedDeletion {
-  // Additional state (if necessary)
+// Additional state (if necessary)
 }
 
 struct UntrackedFileChange {
-  1: ContentId content_id,
+  1: ContentId content_id;
-  2: FileType file_type,
+  2: FileType file_type;
-  3: i64 size,
+  3: i64 size;
 }
 
 struct FileChange {
-  1: required ContentId content_id,
+  1: required ContentId content_id;
-  2: FileType file_type,
+  2: FileType file_type;
   // size is a u64 stored as an i64
-  3: required i64 size,
+  3: required i64 size;
-  4: optional CopyInfo copy_from,
+  4: optional CopyInfo copy_from;
 }
 
 // This is only used optionally so it is OK to use `required` here.
 struct CopyInfo {
-  1: required MPath file,
+  1: required MPath file;
   // cs_id must match one of the parents specified in BonsaiChangeset
-  2: required ChangesetId cs_id,
+  2: required ChangesetId cs_id;
 }
 
 struct FileUnode {
-  1: list<FileUnodeId> parents,
+  1: list<FileUnodeId> parents;
-  2: ContentId content_id,
+  2: ContentId content_id;
-  3: FileType file_type,
+  3: FileType file_type;
-  4: MPathHash path_hash,
+  4: MPathHash path_hash;
-  5: ChangesetId linknode,
+  5: ChangesetId linknode;
 }
 
 union UnodeEntry {
-  1: FileUnodeId File,
+  1: FileUnodeId File;
-  2: ManifestUnodeId Directory,
+  2: ManifestUnodeId Directory;
 }
 
 struct ManifestUnode {
-  1: list<ManifestUnodeId> parents,
+  1: list<ManifestUnodeId> parents;
-  2: map<MPathElement, UnodeEntry> (rust.type = "sorted_vector_map::SortedVectorMap") subentries,
+  2: map<MPathElement, UnodeEntry> (
-  3: ChangesetId linknode,
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) subentries;
+  3: ChangesetId linknode;
 }
 
 struct DeletedManifest {
-  1: optional ChangesetId linknode,
+  1: optional ChangesetId linknode;
-  2: map<MPathElement, DeletedManifestId> (rust.type = "sorted_vector_map::SortedVectorMap") subentries,
+  2: map<MPathElement, DeletedManifestId> (
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) subentries;
 }
 
 struct FsnodeFile {
-  1: ContentId content_id,
+  1: ContentId content_id;
-  2: FileType file_type,
+  2: FileType file_type;
   // size is a u64 stored as an i64
-  3: i64 size,
+  3: i64 size;
-  4: Sha1 content_sha1,
+  4: Sha1 content_sha1;
-  5: Sha256 content_sha256,
+  5: Sha256 content_sha256;
 }
 
 struct FsnodeDirectory {
-  1: FsnodeId id,
+  1: FsnodeId id;
-  2: FsnodeSummary summary,
+  2: FsnodeSummary summary;
 }
 
 struct FsnodeSummary {
-  1: Sha1 simple_format_sha1,
+  1: Sha1 simple_format_sha1;
-  2: Sha256 simple_format_sha256,
+  2: Sha256 simple_format_sha256;
   // Counts and sizes are u64s stored as i64s
-  3: i64 child_files_count,
+  3: i64 child_files_count;
-  4: i64 child_files_total_size,
+  4: i64 child_files_total_size;
-  5: i64 child_dirs_count,
+  5: i64 child_dirs_count;
-  6: i64 descendant_files_count,
+  6: i64 descendant_files_count;
-  7: i64 descendant_files_total_size,
+  7: i64 descendant_files_total_size;
 }
 
 union FsnodeEntry {
-  1: FsnodeFile File,
+  1: FsnodeFile File;
-  2: FsnodeDirectory Directory,
+  2: FsnodeDirectory Directory;
 }
 
 // Content-addressed manifest, with metadata useful for filesystem
@@ -282,33 +296,35 @@ union FsnodeEntry {
 // files and manifests, and the number of files and sub-directories within
 // directories.
 struct Fsnode {
-  1: map<MPathElement, FsnodeEntry> (rust.type = "sorted_vector_map::SortedVectorMap") subentries,
+  1: map<MPathElement, FsnodeEntry> (
-  2: FsnodeSummary summary,
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) subentries;
+  2: FsnodeSummary summary;
 }
 
 struct SkeletonManifestDirectory {
-  1: SkeletonManifestId id,
+  1: SkeletonManifestId id;
-  2: SkeletonManifestSummary summary,
+  2: SkeletonManifestSummary summary;
 }
 
 struct SkeletonManifestSummary {
-  1: i64 child_files_count,
+  1: i64 child_files_count;
-  2: i64 child_dirs_count,
+  2: i64 child_dirs_count;
-  3: i64 descendant_files_count,
+  3: i64 descendant_files_count;
-  4: i64 descendant_dirs_count,
+  4: i64 descendant_dirs_count;
-  5: i32 max_path_len,
+  5: i32 max_path_len;
-  6: i32 max_path_wchar_len,
+  6: i32 max_path_wchar_len;
-  7: bool child_case_conflicts,
+  7: bool child_case_conflicts;
-  8: bool descendant_case_conflicts,
+  8: bool descendant_case_conflicts;
-  9: bool child_non_utf8_filenames,
+  9: bool child_non_utf8_filenames;
-  10: bool descendant_non_utf8_filenames,
+  10: bool descendant_non_utf8_filenames;
-  11: bool child_invalid_windows_filenames,
+  11: bool child_invalid_windows_filenames;
-  12: bool descendant_invalid_windows_filenames,
+  12: bool descendant_invalid_windows_filenames;
 }
 
 struct SkeletonManifestEntry {
   // Present if this is a directory, absent for a file.
-  1: optional SkeletonManifestDirectory directory,
+  1: optional SkeletonManifestDirectory directory;
 }
 
 // Structure-addressed manifest, with metadata useful for traversing manifest
@@ -319,8 +335,10 @@ struct SkeletonManifestEntry {
 // represented by a single skeleton manifest.  Skeleton manifest identities
 // change when files are added or removed.
 struct SkeletonManifest {
-  1: map<MPathElement, SkeletonManifestEntry> (rust.type = "sorted_vector_map::SortedVectorMap") subentries,
+  1: map<MPathElement, SkeletonManifestEntry> (
-  2: SkeletonManifestSummary summary,
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) subentries;
+  2: SkeletonManifestSummary summary;
 }
 
 // Structure that holds a commit graph, usually a history of a file
@@ -361,16 +379,16 @@ struct SkeletonManifest {
 // Note that offset might point to a commit in a next FastlogBatch or even
 // point to batch outside of all previous_batches.
 struct FastlogBatch {
-  1: list<CompressedHashAndParents> latest,
+  1: list<CompressedHashAndParents> latest;
-  2: list<FastlogBatchId> previous_batches,
+  2: list<FastlogBatchId> previous_batches;
 }
 
 typedef i32 ParentOffset (rust.newtype)
 
 struct CompressedHashAndParents {
-  1: ChangesetId cs_id,
+  1: ChangesetId cs_id;
   # Offsets can be negative!
-  2: list<ParentOffset> parent_offsets,
+  2: list<ParentOffset> parent_offsets;
 }
 
 typedef i32 BlameChangeset (rust.newtype)
@@ -384,21 +402,21 @@ enum BlameRejected {
 // Blame V1
 
 struct BlameRange {
-  1: i32 length,
+  1: i32 length;
-  2: ChangesetId csid,
+  2: ChangesetId csid;
-  3: BlamePath path,
+  3: BlamePath path;
   // offset of this range in the origin file (file that introduced this change)
-  4: i32 origin_offset,
+  4: i32 origin_offset;
 }
 
 struct Blame {
-  1: list<BlameRange> ranges,
+  1: list<BlameRange> ranges;
-  2: list<MPath> paths,
+  2: list<MPath> paths;
 }
 
 union BlameMaybeRejected {
-  1: Blame Blame (py3.name = "blame"),
+  1: Blame Blame (py3.name = "blame");
-  2: BlameRejected Rejected,
+  2: BlameRejected Rejected;
 }
 
 // Blame V2
@@ -406,16 +424,16 @@ union BlameMaybeRejected {
 struct BlameRangeV2 {
   // Length (in lines) of this range.  The offset of a range is implicit from
   // the sum of the lengths of the prior ranges.
-  1: i32 length,
+  1: i32 length;
 
   // Index into csids of the changeset that introduced these lines.
-  2: BlameChangeset csid_index,
+  2: BlameChangeset csid_index;
 
   // Index into paths of the path of this file when this line was introduced.
-  3: BlamePath path_index,
+  3: BlamePath path_index;
 
   // The offset of this range at the time that this line was introduced.
-  4: i32 origin_offset,
+  4: i32 origin_offset;
 
   // "Skip past this change" support.
   //
@@ -464,16 +482,16 @@ struct BlameRangeV2 {
 
   // The offset of this range in the file before this range was introduced that
   // was replaced by this range.  Not present for root commits.
-  5: optional i32 parent_offset,
+  5: optional i32 parent_offset;
 
   // The length of the range in the file before this range was introduced that
   // was replaced by this range.  Not present for root commits.
-  6: optional i32 parent_length,
+  6: optional i32 parent_length;
 
   // If this file was being renamed when this line was introduced, this is
   // the index into paths of the original path.  Not present for root commits
   // or if the file has the same name as path_index.
-  7: optional BlamePath renamed_from_path_index,
+  7: optional BlamePath renamed_from_path_index;
 
   // If this is a merge commit, and the file is not in the first parent, then
   // this is the index of the first parent that contains the file that contains
@@ -484,13 +502,13 @@ struct BlameRangeV2 {
   //
   // Note that this is an index into the list of parents in the bonsai
   // changeset, and *not* an index into csids.
-  8: optional i32 parent_index,
+  8: optional i32 parent_index;
 }
 
 struct BlameDataV2 {
   // A list of ranges that describe when the lines of this file were
   // introduced.
-  1: list<BlameRangeV2> ranges,
+  1: list<BlameRangeV2> ranges;
 
   // A mapping of integer indexes to changeset IDs that is used to reduce the
   // repetition of data in ranges.
@@ -503,30 +521,32 @@ struct BlameDataV2 {
   // Changesets are removed from this map when all lines that were added in the
   // changeset are moved and none of the ranges reference it.  This means there
   // are gaps in this mapping, and so a map is used.
-  2: map<i32, ChangesetId> (rust.type = "sorted_vector_map::SortedVectorMap") csids,
+  2: map<i32, ChangesetId> (
+    rust.type = "sorted_vector_map::SortedVectorMap",
+  ) csids;
 
   // The maximum index that is assigned to a changeset id.  This is also the
   // index that would be assigned to the current changeset, as long as the
   // changeset adds new lines.  If the changeset only deletes or merges lines,
   // then this index will not appear in the csids map.
-  3: BlameChangeset max_csid_index,
+  3: BlameChangeset max_csid_index;
 
   // The list of paths that this file has been located at.  This is used to
   // reduce repetition of data in ranges.  Since files are not often moved, and
   // for simplicity, this includes all paths the file has ever been located at,
   // even if they are no longer referenced by any of the ranges.
-  4: list<MPath> paths,
+  4: list<MPath> paths;
 }
 
 union BlameV2 {
   // This version of the file contains full blame information.
-  1: BlameDataV2 full_blame,
+  1: BlameDataV2 full_blame;
 
   // This version of the file was rejected for blaming.
-  2: BlameRejected rejected,
+  2: BlameRejected rejected;
 }
 
 struct RedactionKeyList {
   // List of keys to be redacted
-  1: list<string> keys,
+  1: list<string> keys;
 }

eden/mononoke/reachabilityindex/if/skiplist.thrift

@@ -8,31 +8,31 @@
 include "eden/mononoke/mononoke_types/if/mononoke_types_thrift.thrift"
 
 struct SkiplistEntry {
-  1: RepoId repo_id,
+  1: RepoId repo_id;
-  2: mononoke_types_thrift.ChangesetId cs_id,
+  2: mononoke_types_thrift.ChangesetId cs_id;
-  3: GenerationNum gen,
+  3: GenerationNum gen;
-  4: SkiplistNodeType node_type,
+  4: SkiplistNodeType node_type;
 }
 
 typedef i32 RepoId (rust.newtype)
 
- // Thrift does not support unsigned, so using i64 here
+// Thrift does not support unsigned, so using i64 here
 typedef i64 GenerationNum (rust.newtype)
 
 struct CommitAndGenerationNumber {
-  1: mononoke_types_thrift.ChangesetId cs_id,
+  1: mononoke_types_thrift.ChangesetId cs_id;
-  2: GenerationNum gen,
+  2: GenerationNum gen;
 }
 
 struct SkipEdges {
-  1: list<CommitAndGenerationNumber> edges,
+  1: list<CommitAndGenerationNumber> edges;
 }
 
 struct ParentEdges {
-  1: list<CommitAndGenerationNumber> edges,
+  1: list<CommitAndGenerationNumber> edges;
 }
 
 union SkiplistNodeType {
-  1: SkipEdges SkipEdges,
+  1: SkipEdges SkipEdges;
-  2: ParentEdges ParentEdges,
+  2: ParentEdges ParentEdges;
 }