acecode/src/session/thread_repair.cpp at master · tmoonlight/acecode · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
#include "thread_repair.hpp"

#include "session_manager.hpp"
#include "../commands/compact.hpp"
#include "../utils/uuid.hpp"

#include <algorithm>
#include <cstddef>

namespace acecode {

namespace {

std::vector<std::size_t> real_user_starts(
    const std::vector<ChatMessage>& messages) {
    std::vector<std::size_t> starts;
    for (std::size_t i = 0; i < messages.size(); ++i) {
        if (is_real_user_message(messages[i])) starts.push_back(i);
    }
    return starts;
}

} // namespace

const char* to_string(ThreadRepairStatus status) {
    switch (status) {
        case ThreadRepairStatus::NoChange: return "noChange";
        case ThreadRepairStatus::Repaired: return "repaired";
        case ThreadRepairStatus::HistoryExhausted: return "historyExhausted";
        case ThreadRepairStatus::Failed: return "failed";
    }
    return "failed";
}

ThreadRepairResult plan_thread_repair(
    const std::vector<ChatMessage>& raw_messages,
    const ThreadRepairOptions& options,
    const SessionLoadDiagnostics& load_diagnostics) {
    ThreadRepairResult result;
    result.load_issues = load_diagnostics;

    auto recovered =
        reconstruct_effective_model_history_with_recovery(raw_messages);
    result.history_issues = recovered.stats;
    result.pre_tokens = estimate_message_tokens(recovered.messages);
    result.replacement_history = recovered.messages;

    const auto starts = real_user_starts(recovered.messages);
    std::size_t keep_from = 0;
    const bool target_requires_prune =
        options.target_tokens > 0 &&
        result.pre_tokens > options.target_tokens;
    while (result.pruned_groups + 1 < static_cast<int>(starts.size())) {
        const bool force_first = options.force_prune_one_group &&
                                 result.pruned_groups == 0;
        const int current_tokens = keep_from == 0
            ? result.pre_tokens
            : estimate_message_tokens(std::vector<ChatMessage>(
                  recovered.messages.begin() +
                      static_cast<std::ptrdiff_t>(keep_from),
                  recovered.messages.end()));
        const bool over_target = options.target_tokens > 0 &&
                                 current_tokens > options.target_tokens;
        if (!force_first && !over_target) break;
        ++result.pruned_groups;
        keep_from = starts[static_cast<std::size_t>(result.pruned_groups)];
    }

    if (keep_from > 0) {
        result.pruned_messages = static_cast<int>(keep_from);
        std::vector<ChatMessage> retained(
            recovered.messages.begin() +
                static_cast<std::ptrdiff_t>(keep_from),
            recovered.messages.end());
        result.replacement_history =
            recover_provider_history(retained).messages;
    }
    result.post_tokens = estimate_message_tokens(result.replacement_history);

    const bool recovery_changed = result.history_issues.changed() ||
                                  result.load_issues.recovered();
    if (result.pruned_groups == 0 && !recovery_changed) {
        const bool cannot_meet_target = target_requires_prune ||
            options.force_prune_one_group;
        result.status = cannot_meet_target
            ? ThreadRepairStatus::HistoryExhausted
            : ThreadRepairStatus::NoChange;
        result.reason = cannot_meet_target
            ? "no removable completed user-turn group remains"
            : "provider history is already consistent";
        return result;
    }

    result.checkpoint.id = generate_uuid();
    result.checkpoint.timestamp = iso_timestamp();
    result.checkpoint.trigger = options.trigger;
    result.checkpoint.summary = "Deterministic thread repair";
    result.checkpoint.messages_compressed = result.pruned_messages;
    result.checkpoint.estimated_tokens_saved =
        (std::max)(0, result.pre_tokens - result.post_tokens);
    result.checkpoint.pre_tokens = result.pre_tokens;
    result.checkpoint.post_tokens = result.post_tokens;
    result.checkpoint.replacement_history = result.replacement_history;
    result.status = ThreadRepairStatus::Repaired;
    result.reason = result.pruned_groups > 0
        ? "old completed user-turn groups were pruned"
        : "provider history structure was recovered";
    return result;
}

ThreadRepairResult apply_thread_repair(
    SessionManager* session_manager,
    std::vector<ChatMessage>& provider_history,
    const ThreadRepairOptions& options,
    const SessionLoadDiagnostics& load_diagnostics) {
    ThreadRepairResult result = plan_thread_repair(
        provider_history, options, load_diagnostics);
    if (!result.repaired()) return result;
    if (!session_manager) {
        result.status = ThreadRepairStatus::Failed;
        result.reason = "active session manager is unavailable";
        return result;
    }
    if (!session_manager->append_compact_checkpoint(result.checkpoint)) {
        result.status = ThreadRepairStatus::Failed;
        result.reason = "failed to append repair checkpoint";
        return result;
    }
    provider_history = result.replacement_history;
    return result;
}

nlohmann::json thread_repair_result_to_json(
    const ThreadRepairResult& result,
    const std::string& thread_id) {
    nlohmann::json issues{
        {"malformedToolCalls", result.history_issues.malformed_tool_calls},
        {"duplicateToolCalls", result.history_issues.duplicate_tool_calls},
        {"synthesizedToolResults", result.history_issues.synthesized_tool_results},
        {"standaloneToolResults", result.history_issues.standalone_tool_results},
        {"unexpectedToolResults", result.history_issues.unexpected_tool_results},
        {"duplicateToolResults", result.history_issues.duplicate_tool_results},
        {"emptyAssistantMessages", result.history_issues.empty_assistant_messages},
        {"malformedRecords", result.load_issues.malformed_complete_records},
        {"ignoredPartialTail", result.load_issues.ignored_partial_tail},
        {"recoveredUnterminatedRecord",
         result.load_issues.recovered_unterminated_record},
    };
    nlohmann::json out{
        {"status", to_string(result.status)},
        {"issues", std::move(issues)},
        {"preTokens", result.pre_tokens},
        {"postTokens", result.post_tokens},
        {"prunedGroups", result.pruned_groups},
        {"prunedMessages", result.pruned_messages},
        {"checkpointId", result.checkpoint.id.empty()
                             ? nlohmann::json(nullptr)
                             : nlohmann::json(result.checkpoint.id)},
        {"reason", result.reason},
    };
    if (!thread_id.empty()) out["threadId"] = thread_id;
    return out;
}

} // namespace acecode